wolne_lektury: check credits against the National Library's catalogue, drop 2 not-free translations, fix 3 translators (v2#109) (#190)

- wolne_lektury: BN catalogue check, drop 2 not-free translations, fix 3 translators (v2#109) (a2fa5fc9a27cbc62d298e47c838591d91020d811)

artifacts/source_findings.md

@@ -3605,55 +3605,57 @@ titles are staff or commissioned (the foundation's rights from them are not publ
 **Investigated** — whether every `wolne_lektury` row carries the author, translator and title Wolne Lektury (WL)
 gives, whether the full licence URL, the credits and the "Książka pochodzi z serwisu Wolne Lektury" line are
 recorded per document, and what the text and the edition behind it are. Results are in
-`data/wolne_lektury/wolne_lektury.attribution.jsonl` (6,076 lines, one per row, in Parquet order) and
+`data/wolne_lektury/wolne_lektury.attribution.jsonl` (6,074 lines, one per row, in Parquet order) and
 `data/wolne_lektury/wolne_lektury.license-evidence.json`; `src/build_wolne_lektury_attribution.py` writes both.
 
-**Verdict** — metadata filled and recorded; the text of 460 rows gets their author's and translator's notes at the
+**Verdict** — metadata filled and recorded; the text of 459 rows gets their author's and translator's notes at the
 end (see "Footnote check"), the rest is unchanged. The builder checks every row's SHA-1 and a whole-table digest of the
-text before the notes, and `src/test_attribution_release.py` re-checks both. No row is excluded. The items under
-"Flagged" need a maintainer decision.
+text before the notes, and `src/test_attribution_release.py` re-checks both. 2 rows that WL marks public domain are dropped because
+their translation is not free (see "BN check"); no other row is excluded. The items under "Flagged" need a maintainer
+decision.
 
 **Method** — the RDF header of each book was read from the top of its WL XML
 (`https://wolnelektury.pl/media/book/xml/<slug>.xml`, Range requests) for the 6,065 distinct first slugs of
-`wolne_lektury.licences.tsv`, plus 4 `/api/` calls (the book list, 5.1 MB, and 3 book records, one for the renamed
+`wolne_lektury.licences.tsv` (then 6,076 rows), plus 4 `/api/` calls (the book list, 5.1 MB, and 3 book records, one for the renamed
 book). 6,117 requests in all, one at a time at least 1.2 s apart, 41.2 MB, no 403, 429 or bot check; robots.txt
 (read first) does not disallow these paths. Headers are parsed as XML with DOCTYPE and entities refused. The
 corrected `author` column is the only change this step makes to the Parquet; the notes (below) are added by
 `src/clean_wolne_lektury.py`.
 
-- **Fields** — `author` is 6,039 of 6,076 rows: WL's `dc:creator` names, "Autorzy oryginału: A; Tłumaczenie: T"
+- **Fields** — `author` is 6,037 of 6,074 rows: WL's `dc:creator` names, "Autorzy oryginału: A; Tłumaczenie: T"
   for a translation (the global_voices form), "Tłumaczenie: T" where WL names a translator and no author (6 rows).
   The 37 empty rows are WL placeholders ("Autor nieznany" 36, "Praca zbiorowa" 1) with no named translator.
-  Translator 841 rows, original author 835, title 6,076. Original title 0: WL's header has no such field.
+  Translator 840 rows, original author 834, title 6,074. Original title 0: WL's header has no such field.
   Pseudonyms and cryptonyms stay as WL writes them; that this is the first-edition form is not established.
-- **Licences** — public domain 4,559 (no URL), LAL 1.3 1,467, CC BY-SA 3.0 PL 34, 3.0 Unported 5, 4.0 11, each with
+- **Licences** — public domain 4,557 (no URL), LAL 1.3 1,467, CC BY-SA 3.0 PL 34, 3.0 Unported 5, 4.0 11, each with
   its full URL and version in `license_url`. WL's own link for all 34 "3.0 PL" documents names the Unported text
   (`/by-sa/3.0/`); the sidecar keeps WL's link in `wl_license_url` next to ours (the 3.0 PL URL, as the label says).
-  The sidecar covers all 6,076 rows, not only the 1,517 under a licence; the credit line and processing note are on
+  The sidecar covers all 6,074 rows, not only the 1,517 under a licence; the credit line and processing note are on
   every line.
 - **Credits** — WL's editor, technical editor, funding and thanks lines and its credit notes are copied per row.
   WL's "editor" is the person who prepared the digital edition (239 distinct names), not the print edition's editor.
 - **Normalisation** — in this repository two steps change a text: trimming leading and trailing whitespace
-  (`build_dynaword.py`) and adding the author's and translator's notes at the end of 460 texts
+  (`build_dynaword.py`) and adding the author's and translator's notes at the end of 459 texts
   (`clean_wolne_lektury.py`); the length, language, OCR and duplicate gates dropped 0 rows; `clean_wolne_lektury.py`
-  dropped 63 whole documents (ELTeC novels) and 2 (no usable licence), and relabelled the rest. No paragraph is
+  dropped 63 whole documents (ELTeC novels), 2 (no usable licence) and 2 (translation not
+  free, see "BN check"), and relabelled the rest. No paragraph is
   removed anywhere (`paragraphs_removed_without_a_mark` 0). What SpeakLeash did before is not documented. WL's footer
   is not in any text (`strip_wl_footer` finds none, no text names WL); who removed it is not established.
 - **Notes and prefaces** — every row was scanned: no WL credit text, no `[przypis` marker. Section headings inside
-  texts: Wstęp 73, Przedmowa 50, Od tłumacza 43, Komentarz 10, Posłowie 10, Słowo wstępne 3, Przypisy 2, Od redakcji,
+  texts: Wstęp 73, Przedmowa 49, Od tłumacza 43, Komentarz 10, Posłowie 10, Słowo wstępne 3, Przypisy 2, Od redakcji,
   Od wydawców and Uwagi wydawcy 1 each. A hand read of the first and last 450 characters of 40 random rows
-  (`random.Random(109)`) found no WL credit or note and 3 rows with print-edition matter (a publisher's address, a
-  list of editions, an introduction). WL's txt header lines open the text: ISBN in 1,120 rows, "tłum." in 839.
-- **Edition basis** — WL's print-edition field is `source_edition` (named for 6,032 of 6,076 rows). We take the texts
+  of the then 6,076 (`random.Random(109)`) found no WL credit or note and 3 rows with print-edition matter (a publisher's address, a
+  list of editions, an introduction). WL's txt header lines open the text: ISBN in 1,118 rows, "tłum." in 837.
+- **Edition basis** — WL's print-edition field is `source_edition` (named for 6,030 of 6,074 rows). We take the texts
   to be WL's modernised versions of those editions; 239 rows keep period spelling by a heuristic (at least 5 old-
   spelling words per 1,000), so the modernisation is not uniform.
 
-**Flagged, nothing excluded**
-- Public-domain rows only (a licensed row has its rightsholder's licence): 2,421 have at least one flag.
-  1,311 name a print edition published after 1955, 58 a critical or scholarly edition, 229 an introduction, notes or
-  commentary, 802 name an editor or selector, 802 a translator, 1 a National Library edition. Row ids for the first
-  two are in the evidence file. The editors' and translators' death years are in neither WL nor this repository, so
-  editors who died after 1955 cannot be listed.
+**Flagged**
+- Public-domain rows only (a licensed row has its rightsholder's licence): 2,419 have at least one flag.
+  1,310 name a print edition published after 1955, 58 a critical or scholarly edition, 229 an introduction, notes or
+  commentary, 802 name an editor or selector, 801 a translator, 1 a National Library edition. Row ids for the first
+  two are in the evidence file. The editors' death years are in neither WL nor this repository, so editors who died
+  after 1955 cannot be listed; the authors' and translators' were compared with BN's where BN gives one ("BN check").
 - Precision, from 15 rows drawn by hand for each of the four flags: two patterns were too loose and are removed. "naukow"
   matched the printing house "Drukarnia Naukowa" (9 rows; critical 67 to 58) and "wyd." matched publishers (editor
   credit 1,850 to 802). "wybór" still matches a title word, so the editor flag stays broad. The fixed patterns were
@@ -3662,7 +3664,7 @@ corrected `author` column is the only change this step makes to the Parquet; the
 - Edition review, 2026-10-20 (`edition_review.reviewed` in the evidence file; pinned by a test): 98 rows read, the 58
   `critical` rows, the 10 "Komentarz" rows and the 30 public-domain rows with no WL public-domain date. No row cut or
   dropped. Details and the open points (3 Konopnicka rows that open with an unsigned prose note, 7 of the 30 less
-  certain, 2 translations by an unnamed translator kept as anonymous) are in the evidence file. The 1,311 rows printed
+  certain, 2 translations by an unnamed translator kept as anonymous) are in the evidence file. The 1,310 rows printed
   after 1955 were not checked for the print's editor.
 - WL's editors added glosses inside the author's and translator's notes (`[...; red. WL]`: a word explained in square
   brackets). By the maintainer's decision (2026-10-20) `strip_glosses()` in `src/wolne_lektury_notes.py` strips every
@@ -3683,11 +3685,11 @@ corrected `author` column is the only change this step makes to the Parquet; the
   released text.
 
 **Footnote check — run, notes restored (2026-10-20).** `src/wolne_lektury_notes.py` read WL's book XML (6,064 files,
-one per row slug; WL serves none for the Korczak row, 404) and wrote `data/wolne_lektury/wolne_lektury.notes.jsonl`
-(pinned by SHA-256; each XML file's hash is in it). 460 rows have an author's (`<pa>`) or translator's (`<pt>`) note:
-8,105 and 10,838 notes. The export had dropped them. By the maintainer's decision `src/clean_wolne_lektury.py` adds them
+one per slug of the then 6,076 rows; WL serves none for the Korczak row, 404) and wrote `data/wolne_lektury/wolne_lektury.notes.jsonl`
+(pinned by SHA-256; each XML file's hash is in it). 459 rows have an author's (`<pa>`) or translator's (`<pt>`) note:
+8,104 and 10,838 notes. The export had dropped them. By the maintainer's decision `src/clean_wolne_lektury.py` adds them
 at the end of each of those texts as "Przypisy autora" and "Przypisy tłumacza", numbered from 1 in each, in document
-order: 8,100 author's notes (337 rows) and 10,838 translator's notes (168 rows), +3,909,520 characters and +1,497,180
+order: 8,099 author's notes (336 rows) and 10,838 translator's notes (168 rows), +3,909,359 characters and +1,497,118
 tokens. WL's editorial notes (`<pe>`, `<pr>`) are not restored, none of their text is inside a restored note, and WL's
 glosses inside the restored notes are stripped (above). Left out:
 2 empty notes and 3 author's notes that the export had kept in the text (`wolne_lektury_2371978`,
@@ -3695,6 +3697,37 @@ glosses inside the restored notes are stripped (above). Left out:
 the translator. Not checked: the Korczak row (no XML) and the second book of `wolne_lektury_2375260`. The ELTeC
 dedup scored the texts without the notes, so `wolne_lektury.eltec_pairs.tsv` is unchanged.
 
+**BN check — run, 2 rows dropped (2026-10-20).** Run on the 6,076 rows of the time together with the 100
+`eltec_pol` novels. Each row's BN records (https://data.bn.org.pl, read only; WL's own
+e-book records left out as circular) were found by the author's surname and the title of the work and of its print
+edition; our author and translators were compared with BN's fields 100, 245$c and 700, and for public-domain rows BN's
+death years with 1955 and with WL's `date.pd`. Every row flagged for an author or translator was read by hand
+(translator not ours 112, translator missing in BN 53, author mismatch 31, death year after 1955 1), the other flags in
+samples (no record 50 of 503, pseudonym 50 of 76, other 48 of 120). Of these 345 row-flags (7 on `eltec_pol` rows), 41 are real errors,
+192 catalogue noise, 36 pseudonyms BN resolves to the same person and 76 cannot tell (1 on `eltec_pol`). `bn_check` in the evidence file has the
+details; by the maintainer's decisions of the same day:
+- Dropped as not free (`NOT_FREE` in `src/wolne_lektury_licences.py`), both marked public domain by WL: Nietzsche's
+  "Wola mocy" (`wolne_lektury_2375754`) is a joint translation by Konrad Drzewiecki (d. 1922) and Stefan Frycz
+  (d. 1960; BN record 1826277), protected to the end of 2030; Wells's "Wehikuł czasu" (`wolne_lektury_2371705`) follows
+  the Ossolineum print of 1985, whose translation Juliusz K. Palczewski, born in 1932, revised (BN record 1216473). The
+  Wells row had 1 restored author's note, hence 459 rows with notes, not 460.
+- Translators corrected (`TRANSLATOR_FIXES` in `src/build_wolne_lektury_attribution.py`): "Leon Urlich" to "Leon
+  Ulrich" (`wolne_lektury_2374887`), Józef Nusbaum added next to Szymon Dickstein (`wolne_lektury_2373112`) and Aniela
+  Zagórska added where WL's header names no translator (`wolne_lektury_2372990`), in the sidecar and in `author`. One
+  print-edition note (`SOURCE_EDITION_NOTES`): for Hebbel's diaries (`wolne_lektury_2374125`) WL's `source_edition` is
+  Hebbel's "Judyta" (BN record 1454550). These 4 rows carry a `metadata_note` in the sidecar.
+- Montaigne's "Próby": BN record 2293540 credits the Latin and Greek quotations of book 1 (PIW 1957) to Tadeusz Sinko
+  (d. 1966) and Edmund Cięglewicz (d. 1928) jointly. Of the 466 translator's notes restored to Księga pierwsza
+  (`wolne_lektury_2376515`), 336 end "tłum. Edmund Cięglewicz", 4 of them on a Greek quotation, and none names Sinko;
+  the notes of books 2 and 3 (862 and 622) give a quotation's source and seldom a translation. Kept as WL credits them,
+  with BN's joint credit recorded; which quotations it covers is not established.
+- `natural_order()` (`src/attribution_common.py`) now keeps a bracket after the name it follows: 21 rows read "Adam
+  [Zofia Trzeszczkowska] M-ski", and all 22 rows with the bracket now read "Adam M-ski [Zofia Trzeszczkowska]".
+- Kept as WL wrote them, since none changes whether a row is free: 25 rows whose WL rights line gives a death year BN
+  gives otherwise (for example "Wacław Rolicz-Lieder zm. 1940", BN 1912, 8 rows) and 6 rows whose WL `date.pd` is
+  earlier than the translator's or author's death allows.
+- Open: 74 cannot-tell row-flags on 69 kept rows, left as WL gives them and listed in v2#109.
+
 **Not established** — the editors of the print editions and their death years; the first-edition form of any author
 name; what SpeakLeash changed and who removed the footer; whether
 `source_edition` is the edition the modernised text actually follows.

data/wolne_lektury/wolne_lektury.attribution.jsonl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8768b1ce641f36a5ab7b41dfc2c97cdee31ed3b3b10ae9cb47ad978e318d8b2
-size 10652694
+oid sha256:3492dae04dbc4793fb8eabca6ff2cf55c09dfa35493988a332d479aeaa925285
+size 10649957

data/wolne_lektury/wolne_lektury.licences.tsv

@@ -118,7 +118,6 @@ wolne_lektury_2371701	weintraub-kolysanka-podziemna	exact	public-domain	Domena p
 wolne_lektury_2371702	weintraub-popiol	exact	public-domain	Domena publiczna - Jerzy Kamil Weintraub zm. 1943	
 wolne_lektury_2371703	weintraub-rosniesz-nocy-kamienna-i-coraz-mi-ciezej	exact	public-domain	Domena publiczna - Jerzy Kamil Weintraub zm. 1943	
 wolne_lektury_2371704	weintraub-wiersze-grudniowe	exact	public-domain	Domena publiczna - Jerzy Kamil Weintraub zm. 1943	
-wolne_lektury_2371705	wells-wehikul-czasu	exact	public-domain	Domena publiczna - Herbert George Wells zm. 1946	
 wolne_lektury_2371706	wergiliusz-eneida	exact	public-domain	Domena publiczna - Tadeusz Karyłowski, zm. 1945.	
 wolne_lektury_2371707	wesele	exact	public-domain	Domena publiczna - Stanisław Wyspiański zm. 1907	
 wolne_lektury_2371708	wesele-figara	exact	public-domain	Domena publiczna - Tadeusz Boy-Żeleński zm. 1941	
@@ -4124,7 +4123,6 @@ wolne_lektury_2375750	niestatek-oczy-sa-ogien-czolo-jest-zwierciadlem	exact	publ
 wolne_lektury_2375751	niestatek-predzej-kto-wiatr-w-wor-zamknie-predzej-i-promieni	exact	public-domain	Domena publiczna - Jan Andrzej Morsztyn zm. 1693	
 wolne_lektury_2375752	nietzsche-jutrzenka-mysli-o-przesadach-moralnych	exact	public-domain	Domena publiczna - Stanisław Wyrzykowski zm. 1949	
 wolne_lektury_2375753	nietzsche-wedrowiec-i-jego-cien	exact	public-domain	Domena publiczna - Konrad Drzewiecki zm. 1922	
-wolne_lektury_2375754	nietzsche-wola-mocy	exact	public-domain	Domena publiczna - Konrad Drzewiecki zm. 1922	
 wolne_lektury_2375755	nietzsche-zmierzch-bozyszcz	exact	public-domain	Domena publiczna - Stanisław Wyrzykowski zm. 1949	
 wolne_lektury_2375756	nie-udalo-sie	exact	public-domain	Domena publiczna - Anton Czechow zm. 1904	
 wolne_lektury_2375757	niewierny	exact	public-domain	Domena publiczna - Kazimierz Przerwa-Tetmajer zm. 1940	

data/wolne_lektury/wolne_lektury.license-evidence.json

@@ -1,14 +1,176 @@
 {
-  "built_from_parquet_sha256": "6b5429ed5dcf8e9bf9404f856ebfcc09bf5f178fe79e8aeffeb1db468dd0501f",
+  "bn_check": {
+    "checked_at": "2026-10-20",
+    "decided_at": "2026-10-20",
+    "dropped_as_not_free": {
+      "wolne_lektury_2371705": "Wells, \"Wehikuł czasu\" (wells-wehikul-czasu): transcribed from the Ossolineum print (Wrocław 1985), whose translation was revised by an editor born in 1932 (BN record 1216473: \"przeł. [z ang.] Feliks Wermiński ; przekł. zweryfikował i oprac. Juliusz K. Palczewski\"; \"Palczewski, Juliusz K. (1932- )\"). Wolne Lektury credits Wermiński (d. 1917) only.",
+      "wolne_lektury_2375754": "Nietzsche, \"Wola mocy\" (nietzsche-wola-mocy): a joint translation by Stefan Frycz and Konrad Drzewiecki (BN record 1826277, Warszawa: J. Mortkowicz, 1911: \"przeł. Stefan Frycz i Konrad Drzewiecki\"; \"Frycz, Stefan (1876-1960)\"). Wolne Lektury counts Drzewiecki alone (d. 1922); a joint work is protected for 70 years after the last co-author's death, so to the end of 2030."
+    },
+    "joint_credit_kept": {
+      "counted": "translator's notes restored at the end of the text; crediting_cieglewicz: those that contain \"tłum. Edmund Cięglewicz\"; of_them_greek: those of them whose quotation WL marks \"(gr.)\"",
+      "finding": "BN record 2293540 (Próby, Ks. 1, Warszawa: PIW, 1957, the edition Wolne Lektury names for all three rows) reads \"[cytaty łac. i grec. przeł. Tadeusz Sinko i Edmund Cięglewicz]\", with \"Cięglewicz, Edmund (1862-1928)\" and \"Sinko, Tadeusz (1877-1966)\". Wolne Lektury's print edition for Księga pierwsza reads \"cytaty łac. tłum. Edmund Cięgielewicz\", its translator's notes end the citation of each quotation they translate with \"tłum. Edmund Cięglewicz\", and none names Sinko. The notes of the other two books give the source of a quotation and seldom a translation. Kept as Wolne Lektury credits them (maintainer decision 2026-10-20); which quotations BN's joint credit covers is not established.",
+      "rows": {
+        "wolne_lektury_2376514": {
+          "book": "Księga druga",
+          "crediting_cieglewicz": 0,
+          "greek_characters": 0,
+          "naming_sinko": 0,
+          "of_them_greek": 0,
+          "translator_notes": 862
+        },
+        "wolne_lektury_2376515": {
+          "book": "Księga pierwsza",
+          "crediting_cieglewicz": 336,
+          "greek_characters": 674,
+          "naming_sinko": 0,
+          "of_them_greek": 4,
+          "translator_notes": 466
+        },
+        "wolne_lektury_2376516": {
+          "book": "Księga trzecia",
+          "crediting_cieglewicz": 0,
+          "greek_characters": 0,
+          "naming_sinko": 0,
+          "of_them_greek": 0,
+          "translator_notes": 622
+        }
+      }
+    },
+    "kept_as_wolne_lektury_wrote": {
+      "death_year_in_wl_rights": {
+        "Bolesław Leśmian zm. 1938": {
+          "bn_death_year": 1937,
+          "ids": [
+            "wolne_lektury_2373317",
+            "wolne_lektury_2373354",
+            "wolne_lektury_2373355"
+          ]
+        },
+        "Jan Kiliński zm. 1760": {
+          "bn_death_year": 1819,
+          "ids": [
+            "wolne_lektury_2374523"
+          ]
+        },
+        "Jerzy Żuławski zm. 1914": {
+          "bn_death_year": 1915,
+          "ids": [
+            "wolne_lektury_2372125"
+          ]
+        },
+        "Wacław Rolicz-Lieder zm. 1940": {
+          "bn_death_year": 1912,
+          "ids": [
+            "wolne_lektury_2376896",
+            "wolne_lektury_2376897",
+            "wolne_lektury_2376898",
+            "wolne_lektury_2376899",
+            "wolne_lektury_2376900",
+            "wolne_lektury_2376901",
+            "wolne_lektury_2376902",
+            "wolne_lektury_2376903"
+          ]
+        },
+        "Władysław Witwicki zm. 1947": {
+          "bn_death_year": 1948,
+          "ids": [
+            "wolne_lektury_2376310",
+            "wolne_lektury_2376311",
+            "wolne_lektury_2376312",
+            "wolne_lektury_2376313",
+            "wolne_lektury_2376314",
+            "wolne_lektury_2376315",
+            "wolne_lektury_2376316",
+            "wolne_lektury_2376317",
+            "wolne_lektury_2376318",
+            "wolne_lektury_2376319",
+            "wolne_lektury_2376321",
+            "wolne_lektury_2376322"
+          ]
+        }
+      },
+      "meaning": "the check found these and the maintainer left them as Wolne Lektury wrote them (2026-10-20); none changes whether a row is free",
+      "wl_date_pd_early": {
+        "wolne_lektury_2372974": "the translator Franciszek Mirandola died 1930 (BN): free from 2001, not 1903",
+        "wolne_lektury_2374125": "the translator Karol Irzykowski died 1944 (BN): free from 2015, not 1981",
+        "wolne_lektury_2374971": "the translator Antoni Lange died 1929 (BN): free from 2000, not 1938",
+        "wolne_lektury_2375897": "the translator Cyprian Bazylik died 1600 (BN): free from 1671, not 1643",
+        "wolne_lektury_2376199": "the author Arthur Conan Doyle died 1930 (BN): free from 2001, not 1994",
+        "wolne_lektury_2376940": "the translator Maria Cumft-Pieńkowska died 1920 (BN): free from 1991, not 1971"
+      }
+    },
+    "method": "each row's BN records found by the author's surname and the title of the work and of its print edition; our author and translators compared with BN's fields 100, 245$c and 700, and for public-domain rows BN's death years with 1955 and with WL's date.pd. Every row flagged for an author or translator was read by hand, the other flags in samples.",
+    "open": "the cannot_tell rows not named in this section are left as Wolne Lektury gives them; they are listed in polish-dynaword-v2#109",
+    "pseudonym_with_real_name": {
+      "meaning": "Zofia Trzeszczkowska wrote as Adam M-ski. Most of WL's headers give \"M-ski, Adam [Zofia Trzeszczkowska]\"; natural_order() keeps the bracket after the name.",
+      "rows": 22,
+      "translator": "Adam M-ski [Zofia Trzeszczkowska]"
+    },
+    "public_domain_rows_with_a_bn_death_after_1955": "one, \"Wola mocy\", dropped (dropped_as_not_free)",
+    "reviewed_by_hand": {
+      "in_full": {
+        "author_mismatch": 31,
+        "death_year_after_1955": 1,
+        "translator_missing_in_bn": 53,
+        "translator_not_ours": 112
+      },
+      "in_samples": {
+        "no_record": "50 of 503",
+        "other": "48 of 120",
+        "pseudonym": "50 of 76"
+      },
+      "labels": {
+        "cannot_tell": 76,
+        "catalogue_noise": 192,
+        "pseudonym_ok": 36,
+        "real_error": 41
+      },
+      "scope": "the check covered the 6,076 Wolne Lektury rows of the time (this table's and the 2 in NOT_FREE) and the 100 eltec_pol novels; 7 of the row-flags read are on eltec_pol rows (5 pseudonym_ok, 1 catalogue_noise, 1 cannot_tell)",
+      "unit": "a row and flag; a row can have several flags"
+    },
+    "source": "the National Library's catalogue, https://data.bn.org.pl (read only; WL's own e-book records left out as circular)",
+    "source_edition_notes": {
+      "wolne_lektury_2374125": "Wolne Lektury's print edition (`source_edition`) is Hebbel's play \"Judyta\" in Kazimierz Kaszewski's translation (BN record 1454550, Warszawa 1881), not this book, Hebbel's diaries in Karol Irzykowski's translation (Wolne Lektury's rights line: \"Karol Irzykowski zm. 1944\"). Which print the text follows is not established."
+    },
+    "translators_fixed": {
+      "wolne_lektury_2372990": {
+        "evidence": "Wolne Lektury's print edition reads \"tłum. Aniela Zagórska\" and its rights line \"Aniela Zagórska zm. 1943\"; BN record 1902875 (Warszawa: PIW, 1957), \"przeł. Aniela Zagórska\", \"Zagórska, Aniela (1881-1943)\".",
+        "ours": [
+          "Aniela Zagórska"
+        ],
+        "wolne_lektury": []
+      },
+      "wolne_lektury_2373112": {
+        "evidence": "Wolne Lektury's print edition reads \"tłum. Szymon Dickstein i Józef Nusbaum\"; BN record 2114549 (Warszawa 1885), \"przeł. Szymon Dickstein i Józef Nusbaum\", \"Nusbaum-Hilarowicz, Józef (1859-1917)\".",
+        "ours": [
+          "Szymon Dickstein",
+          "Józef Nusbaum"
+        ],
+        "wolne_lektury": [
+          "Szymon Dickstein"
+        ]
+      },
+      "wolne_lektury_2374887": {
+        "evidence": "Wolne Lektury's own rights line reads \"Leon Ulrich zm. 1885\"; BN record 1504075 (Kraków: Gebethner, 1895), \"przekład J. Ulricha\", \"Ulrich, Leon (1811-1885)\".",
+        "ours": [
+          "Leon Ulrich"
+        ],
+        "wolne_lektury": [
+          "Leon Urlich"
+        ]
+      }
+    }
+  },
+  "built_from_parquet_sha256": "cae78e0ea34138934839225c39bbb895c151a22a4be0f5b11ddd29d56df2d568",
   "edition_review": {
     "flags": {
       "apparatus": 229,
       "critical": 58,
       "editor_credit": 802,
       "national_library": 1,
-      "print_after_1955": 1311,
-      "translation": 587,
-      "translator_credited": 802
+      "print_after_1955": 1310,
+      "translation": 585,
+      "translator_credited": 801
     },
     "ids": {
       "critical": [
@@ -101,7 +263,6 @@
         "wolne_lektury_2371702",
         "wolne_lektury_2371703",
         "wolne_lektury_2371704",
-        "wolne_lektury_2371705",
         "wolne_lektury_2371707",
         "wolne_lektury_2371708",
         "wolne_lektury_2371773",
@@ -1509,7 +1670,7 @@
       ],
       "not_checked": {
         "reason": "the rows printed after 1955 (`edition_review.ids.print_after_1955`): the print's editor was not checked.",
-        "rows": 1311,
+        "rows": 1310,
         "wl_date_pd_wrong": {
           "wolne_lektury_2374660": "WL's date.pd 2012 is wrong (Konopnicka died 1910); the row is public domain either way."
         }
@@ -1518,7 +1679,7 @@
       "reviewed_at": "2026-10-20",
       "rows": 98
     },
-    "rows_flagged": 2421,
+    "rows_flagged": 2419,
     "scope": "public-domain rows; a licensed row has its rightsholder's licence",
     "wl_editor_credits": {
       "distinct_names": 239,
@@ -1526,7 +1687,7 @@
       "most_common": [
         [
           "Aleksandra Sekuła",
-          2211
+          2210
         ],
         [
           "Paweł Kozioł",
@@ -1534,11 +1695,11 @@
         ],
         [
           "Wojciech Kotwica",
-          669
+          668
         ],
         [
           "Dorota Kowalska",
-          574
+          573
         ],
         [
           "Weronika Trzeciak",
@@ -1599,12 +1760,12 @@
       "wolne_lektury_2375260": "its second slug has no XML here; only the first book's XML was read"
     },
     "notes_file": "wolne_lektury.notes.jsonl",
-    "notes_file_sha256": "07fe95a8265f6b40512305cf39b4960fbd16ec2ba9c4c68f7ce92b681d5a51a1",
+    "notes_file_sha256": "da18d6a7eebf88f63efbe1e082d5e2cdc387bf9bbf808cc9d0a6411dbf14d8e1",
     "restored": {
       "pa": {
-        "characters": 1854770,
-        "notes": 8100,
-        "rows": 337
+        "characters": 1854631,
+        "notes": 8099,
+        "rows": 336
       },
       "pt": {
         "characters": 1923101,
@@ -1616,7 +1777,7 @@
       "pa": "\"Przypisy autora\", then the notes numbered from 1",
       "pt": "\"Przypisy tłumacza\", then the notes numbered from 1"
     },
-    "rows_changed": 460,
+    "rows_changed": 459,
     "source": "Wolne Lektury's book XML, one file per row slug (https://wolnelektury.pl/media/book/xml/<slug>.xml), read 2026-10-20",
     "status": "restored",
     "wl_glosses_stripped": {
@@ -5478,6 +5639,12 @@
       "stage": "src/clean_wolne_lektury.py",
       "step": "drop a whole document for which Wolne Lektury gives no usable licence"
     },
+    {
+      "changes_text": "no",
+      "rows_dropped": 2,
+      "stage": "src/clean_wolne_lektury.py",
+      "step": "drop a whole document that Wolne Lektury marks public domain although its translation is not free (bn_check below)"
+    },
     {
       "changes_text": "no",
       "stage": "src/clean_wolne_lektury.py",
@@ -5496,7 +5663,7 @@
   ],
   "notes_and_prefaces": {
     "hand_sample": {
-      "draw": "random.Random(109).sample(ids, 40)",
+      "draw": "random.Random(109).sample(ids, 40), ids: the 6,076 rows of the time in id order (this table's and the 2 in NOT_FREE, neither drawn)",
       "read": "the first and last 450 characters of each text",
       "result": {
         "from_the_print_edition": {
@@ -5524,21 +5691,21 @@
   "source": "wolne_lektury",
   "summary": {
     "author_empty": 37,
-    "author_filled": 6039,
+    "author_filled": 6037,
     "author_missing_in_wl": 0,
     "author_placeholder": 37,
-    "edition_flag_rows": 2421,
+    "edition_flag_rows": 2419,
     "edition_flags": {
       "apparatus": 229,
       "critical": 58,
       "editor_credit": 802,
       "national_library": 1,
-      "print_after_1955": 1311,
-      "translation": 587,
-      "translator_credited": 802
+      "print_after_1955": 1310,
+      "translation": 585,
+      "translator_credited": 801
     },
     "license_url_given_by_wl": {
-      "": 4577,
+      "": 4575,
       "Domena publiczna": 4,
       "Licencja Wolnej Sztuki 1.3.": 1,
       "http://artlibre.org/licence/lal/pl": 10,
@@ -5554,21 +5721,21 @@
       "CC-BY-SA-3.0-PL": 34,
       "CC-BY-SA-4.0": 11,
       "LAL-1.3": 1467,
-      "public-domain": 4559
+      "public-domain": 4557
     },
-    "original_author_filled": 835,
+    "original_author_filled": 834,
     "original_title_filled": 0,
-    "print_after_1955_and_public_domain": 1311,
-    "rows": 6076,
-    "source_edition_filled": 6032,
-    "title_filled": 6076,
-    "translator_filled": 841,
+    "print_after_1955_and_public_domain": 1310,
+    "rows": 6074,
+    "source_edition_filled": 6030,
+    "title_filled": 6074,
+    "translator_filled": 840,
     "wl_url_names_another_version": 34
   },
   "text_checks": {
     "footer_found_by_strip_wl_footer": 0,
-    "header_isbn_line": 1120,
-    "header_translator_line": 839,
+    "header_isbn_line": 1118,
+    "header_translator_line": 837,
     "leading_or_trailing_whitespace": 0,
     "period_spelling": {
       "ids": [
@@ -5815,7 +5982,7 @@
       "rows": 239,
       "words_per_1000_at_least": 5.0
     },
-    "rows": 6076,
+    "rows": 6074,
     "section_headings": {
       "komentarz": {
         "ids": [
@@ -5941,7 +6108,6 @@
           "wolne_lektury_2375693",
           "wolne_lektury_2375752",
           "wolne_lektury_2375753",
-          "wolne_lektury_2375754",
           "wolne_lektury_2375755",
           "wolne_lektury_2375868",
           "wolne_lektury_2375870",
@@ -5960,7 +6126,7 @@
           "wolne_lektury_2377379",
           "wolne_lektury_2377381"
         ],
-        "rows": 50
+        "rows": 49
       },
       "przypisy": {
         "ids": [
@@ -6089,7 +6255,7 @@
     "wolne_lektury_credit_text": 1,
     "wolne_lektury_footnote_markers": 5
   },
-  "text_digest": "a0d669c4c83fc9c624bc48eb610773d7d363f152523630a196990184527df8ab",
+  "text_digest": "9fcc2c3e8e5a7bb3b80bc61fcb086a7148613144bde1adfc067bb6e92afa2b23",
   "wl_glosses_in_main_text": {
     "characters": 90,
     "glosses": 2,

data/wolne_lektury/wolne_lektury.md

@@ -11,30 +11,30 @@ Wolne Lektury (school readings)
 - **Added:** 2026-06-15
 
 ## Licensing — traceable basis
-Each document's `license` is the one its own Wolne Lektury book states in dc:rights / dc:rights.license, never its parent collection's (src/wolne_lektury_licences.py, polish-dynaword-v2#57). Documents are matched to books by text: the same head and tail, or, where Wolne Lektury has edited a text since the scrape, a 5-word-shingle share of at least 0.6 at both ends; 17 were matched by hand. A label is set only when the rights text is unambiguous: a public-domain statement with no year after 1955, or one licence with its version. wolne_lektury.licences.tsv lists each document's book (its slug on wolnelektury.pl), the Wolne Lektury rights text and the label. wolne_lektury.attribution.jsonl has one line per document, in the order of the Parquet: title, author, translator, the link to the work on wolnelektury.pl, the full licence URL with its version (CC BY-SA 3.0 PL, 3.0 Unported and 4.0 are kept apart; Wolne Lektury's own link for each of the 34 "3.0 PL" documents names the Unported text and is kept next to ours), every person and note Wolne Lektury credits, the print edition and publisher it names, the line "Książka pochodzi z serwisu Wolne Lektury" and a note on how the text was processed. The released text is Wolne Lektury's plain-text export without its footer, with the author's and translator's notes of the book added at the end where it has them; the footer's content is restored in the sidecar. Our assessment is that these are Wolne Lektury's modernised versions of the print editions, not the editions as first printed. The edition a text follows is the one in Wolne Lektury's print-edition field, kept in the sidecar as `source_edition` (named for 6,032 of the 6,076 documents). Wolne Lektury's editorial notes are not taken over (see the scope note).
+Each document's `license` is the one its own Wolne Lektury book states in dc:rights / dc:rights.license, never its parent collection's (src/wolne_lektury_licences.py, polish-dynaword-v2#57). Documents are matched to books by text: the same head and tail, or, where Wolne Lektury has edited a text since the scrape, a 5-word-shingle share of at least 0.6 at both ends; 17 were matched by hand. A label is set only when the rights text is unambiguous: a public-domain statement with no year after 1955, or one licence with its version. wolne_lektury.licences.tsv lists each document's book (its slug on wolnelektury.pl), the Wolne Lektury rights text and the label. wolne_lektury.attribution.jsonl has one line per document, in the order of the Parquet: title, author, translator, the link to the work on wolnelektury.pl, the full licence URL with its version (CC BY-SA 3.0 PL, 3.0 Unported and 4.0 are kept apart; Wolne Lektury's own link for each of the 34 "3.0 PL" documents names the Unported text and is kept next to ours), every person and note Wolne Lektury credits, the print edition and publisher it names, the line "Książka pochodzi z serwisu Wolne Lektury" and a note on how the text was processed. The released text is Wolne Lektury's plain-text export without its footer, with the author's and translator's notes of the book added at the end where it has them; the footer's content is restored in the sidecar. Our assessment is that these are Wolne Lektury's modernised versions of the print editions, not the editions as first printed. The edition a text follows is the one in Wolne Lektury's print-edition field, kept in the sidecar as `source_edition` (named for 6,030 of the 6,074 documents). Wolne Lektury's editorial notes are not taken over (see the scope note).
 
 ## Provenance
-Pulled from SpeakLeash's public redistribution (`speakleash-ds-pub`, key `wolne_lektury_corpus`) of Wolne Lektury's plain-text export. What SpeakLeash did to the text is not documented, and who removed the export's footer is not established. In this repository a text is changed in two ways: leading and trailing whitespace is trimmed, and the 460 texts whose Wolne Lektury book has author's or translator's notes get those notes as sections at the end ("Przypisy autora", "Przypisy tłumacza"; Wolne Lektury's editorial notes are not included; src/wolne_lektury_notes.py). No paragraph is removed, and 63 whole documents were dropped as eltec_pol novels and 2 for want of a usable licence (src/clean_wolne_lektury.py). Wolne Lektury's metadata, fetched from its public API and book XML, fills `author` ("Autorzy oryginału: A; Tłumaczenie: T" for a translation) and the sidecar; src/build_wolne_lektury_attribution.py writes both and wolne_lektury.license-evidence.json, and checks every text against its SHA-1.
+Pulled from SpeakLeash's public redistribution (`speakleash-ds-pub`, key `wolne_lektury_corpus`) of Wolne Lektury's plain-text export. What SpeakLeash did to the text is not documented, and who removed the export's footer is not established. In this repository a text is changed in two ways: leading and trailing whitespace is trimmed, and the 459 texts whose Wolne Lektury book has author's or translator's notes get those notes as sections at the end ("Przypisy autora", "Przypisy tłumacza"; Wolne Lektury's editorial notes are not included; src/wolne_lektury_notes.py). No paragraph is removed, and 63 whole documents were dropped as eltec_pol novels, 2 for want of a usable licence and 2 whose translation is not free (src/clean_wolne_lektury.py). Wolne Lektury's metadata, fetched from its public API and book XML, fills `author` ("Autorzy oryginału: A; Tłumaczenie: T" for a translation) and the sidecar; src/build_wolne_lektury_attribution.py writes both and wolne_lektury.license-evidence.json, and checks every text against its SHA-1.
 
 ## Statistics
 | documents | characters | tokens (tiktoken proxy) |
 |---:|---:|---:|
-| 6,076 | 234,671,405 | 92,151,055 |
+| 6,074 | 233,812,331 | 91,823,395 |
 
 
 ## Per-document license metadata
 | license | documents |
 |---|---:|
-| `public-domain` | 4,559 |
+| `public-domain` | 4,557 |
 | `LAL-1.3` | 1,467 |
 | `CC-BY-SA-3.0-PL` | 34 |
 | `CC-BY-SA-4.0` | 11 |
 | `CC-BY-SA-3.0` | 5 |
 
-Author metadata present for **6,039** documents. Empty values mean the upstream record did not expose a machine-readable author field.
+Author metadata present for **6,037** documents. Empty values mean the upstream record did not expose a machine-readable author field.
 
 ## Legal scope note
-For v0.2.6, 63 documents whose text is an eltec_pol novel are dropped (share of a document's 5-word shingles found in one ELTeC novel >= 0.9; src/clean_wolne_lektury.py, polish-dynaword-v2#72). eltec_pol keeps those 50 novels whole under CC BY 4.0. Three documents only partly in an ELTeC novel are kept: both tomes of Mniszek's Trędowata (the ELTeC text is shorter) and Kochanowski's Pieśń IX, quoted in Łoziński's Oko proroka. Every pair is listed in wolne_lektury.eltec_pairs.tsv. 2 documents are also dropped because Wolne Lektury gives them no usable licence: two Horace odes translated by Paweł Kozioł, one marked only "CC BY-SA" with no version, the other public domain for Marceli Motty's translation and "CC BY-SA" for Kozioł's (polish-dynaword-v2#57). Flagged, nothing excluded (polish-dynaword-v2#109). `author` is Wolne Lektury's name form (pseudonyms and cryptonyms as Wolne Lektury gives them); the first-edition form of a name is not established. 37 documents have no author because Wolne Lektury gives only "Autor nieznany" or "Praca zbiorowa"; Wolne Lektury has no field for the original title of a translation. Of the public-domain documents, 1,311 name a print edition published after 1955, 58 a critical edition and 229 an introduction, notes or commentary; the editors of those editions and their death years are not established. 239 documents keep period spelling by our heuristic count, so the modernisation is not uniform. The text of 10 documents (all under LAL 1.3, no print edition named) has a "Komentarz" heading, which we read as Wolne Lektury's own commentary. In a hand check of 40 random documents, 3 carried matter from the print edition (a publisher's address, a list of editions, an introduction) and none carried Wolne Lektury's credits or notes. The export had no footnotes. From Wolne Lektury's book XML (read 2026-10-20) the author's and translator's notes are restored at the end of the 460 texts that have them: 8,100 author's notes under "Przypisy autora" in 337 texts and 10,838 translator's notes under "Przypisy tłumacza" in 168 texts. Wolne Lektury's editorial notes are not restored, and the 668 glosses its editors added inside 632 of the restored notes (in square brackets, signed "red. WL", "WL" or the like; 131 texts, 77,065 characters) are stripped from them; brackets without that signature are kept. The two such glosses in the main text of two documents (90 characters) are stripped too. Two empty notes and three that the export had kept in the text are left out. Wolne Lektury serves no XML for one document (Korczak) and the second book of one merged document was not read, so their notes are not checked. Every public-domain document with translator's notes has its book's public-domain date, which counts the translator.
+For v0.2.6, 63 documents whose text is an eltec_pol novel are dropped (share of a document's 5-word shingles found in one ELTeC novel >= 0.9; src/clean_wolne_lektury.py, polish-dynaword-v2#72). eltec_pol keeps those 50 novels whole under CC BY 4.0. Three documents only partly in an ELTeC novel are kept: both tomes of Mniszek's Trędowata (the ELTeC text is shorter) and Kochanowski's Pieśń IX, quoted in Łoziński's Oko proroka. Every pair is listed in wolne_lektury.eltec_pairs.tsv. 2 documents are also dropped because Wolne Lektury gives them no usable licence: two Horace odes translated by Paweł Kozioł, one marked only "CC BY-SA" with no version, the other public domain for Marceli Motty's translation and "CC BY-SA" for Kozioł's (polish-dynaword-v2#57). 2 more are dropped after a check of Wolne Lektury's credits against the National Library's catalogue (polish-dynaword-v2#109): Wolne Lektury marks both public domain, but Nietzsche's "Wola mocy" is a joint translation by Konrad Drzewiecki and Stefan Frycz (d. 1960), and Wells's "Wehikuł czasu" follows a 1985 print whose translation an editor born in 1932 revised. The same check corrected the translators of 3 documents and found that the print edition named for a fourth (Hebbel's diaries) is another book; these 4 carry a `metadata_note` in the sidecar. In Montaigne's Próby (3 documents) Wolne Lektury's translator's notes credit the translated quotations to Edmund Cięglewicz, while the National Library's record of the print edition credits Tadeusz Sinko (d. 1966) and Cięglewicz jointly; which quotations each translated is not established. Wrong death years in the rights lines of 25 documents and early public-domain dates of 6 are kept as Wolne Lektury gives them; none changes whether a document is free. Otherwise flagged, nothing more excluded (polish-dynaword-v2#109). `author` is Wolne Lektury's name form (pseudonyms and cryptonyms as Wolne Lektury gives them); the first-edition form of a name is not established. 37 documents have no author because Wolne Lektury gives only "Autor nieznany" or "Praca zbiorowa"; Wolne Lektury has no field for the original title of a translation. Of the public-domain documents, 1,310 name a print edition published after 1955, 58 a critical edition and 229 an introduction, notes or commentary; the editors of those editions and their death years are not established. 239 documents keep period spelling by our heuristic count, so the modernisation is not uniform. The text of 10 documents (all under LAL 1.3, no print edition named) has a "Komentarz" heading, which we read as Wolne Lektury's own commentary. In a hand check of 40 random documents, 3 carried matter from the print edition (a publisher's address, a list of editions, an introduction) and none carried Wolne Lektury's credits or notes. The export had no footnotes. From Wolne Lektury's book XML (read 2026-10-20) the author's and translator's notes are restored at the end of the 459 texts that have them: 8,099 author's notes under "Przypisy autora" in 336 texts and 10,838 translator's notes under "Przypisy tłumacza" in 168 texts. Wolne Lektury's editorial notes are not restored, and the 668 glosses its editors added inside 632 of the restored notes (in square brackets, signed "red. WL", "WL" or the like; 131 texts, 77,065 characters) are stripped from them; brackets without that signature are kept. The two such glosses in the main text of two documents (90 characters) are stripped too. Two empty notes and three that the export had kept in the text are left out. Wolne Lektury serves no XML for one document (Korczak) and the second book of one merged document was not read, so their notes are not checked. Every public-domain document with translator's notes has its book's public-domain date, which counts the translator.
 
 Statistics were recomputed directly from the released parquet file.
 
@@ -44,7 +44,7 @@ Minimal, per Dynaword guidelines (heavy filtering left to downstream use):
 - drop non-Polish (diacritic ratio): **0**
 - exact cross-source dedup (sha1): **0**
 - OCR alpha-ratio < 0.70 (OCR sources only): **0**
-- read 6,141 → kept 6,076
+- read 6,141 → kept 6,074
 
 Token counts are a fast tiktoken (cl100k) proxy (~1% off Llama-3); the canonical
 Llama-3 count is computed at release.

data/wolne_lektury/wolne_lektury.notes.jsonl

@@ -21,7 +21,6 @@
 {"date_pd": "1946", "id": "wolne_lektury_2371680", "pa": ["com w tym samym przedmiocie mówiąc o ludach germańskich, uważanych w spełnianiu dziejowego posłannictwa swego przed trzydziestu laty powiedział --- zob. [Karol Libelt,] Pisma pomniejsze tom III strona 101.", "Jeżeli nas poznaki nie mylą, a wskazują na nie zewsząd mężowie, przenikający czas rozumem i natchnieniem, to nowa idea się rodzi i wstępuje w świat --- zob. Ojcze Nasz Augusta Cieszkowskiego."], "pt": [], "slug": "wcielajace-sie-idee-czasu", "xml_sha256": "eec9efb01e69d70cb94402abb1eb5486600b9b86949c203320aa9e2d06185c12"}
 {"date_pd": "1995", "id": "wolne_lektury_2371692", "pa": [], "pt": ["prefekt Worms-Clavelin administrował powściągliwie, z placem Beauvau korespondował zwięźle --- przy placu Beauvau w Paryżu mieściło się ministerstwo spraw wewnętrznych."], "slug": "w-cieniu-wiazow", "xml_sha256": "641927b2273a106c8e14c38f88eaf4e749196c4ce2e9219275bcdfa98e3c8a62"}
 {"date_pd": "2012", "id": "wolne_lektury_2371694", "pa": [], "pt": ["księżnej (...) diuszessy --- polski język nie ma terminów na odróżnienie tytułów duc i prince, które trzeba nam tłumaczyć jednakowo przez książę. Toteż w dalszym ciągu nieraz przyjdzie nam się spotkać z dwoma książętami de Guermantes: Błażejem (duc de Guermantes) i Gilbertem (prince de Guermantes), oraz z dwiema księżnymi: Błażejową (Oriona) i Gilbertową (Maria). Na ogół duc stoi wyżej w hierarchii tytułów, poza paroma rodami, w których tytuł prince posiadał szczególne dostojeństwo, oraz poza ,,książętami krwi\", członkami panujących domów, którym również daje się ogólnie tytuł prince, ale barwiący się wówczas innym odcieniem. Książę Orleański, duc de Chartres, hrabia Paryża, potomkowie francuskiej rodziny panującej, wciąż przez rojalistów uznawanej za królewską, mają inny porządek tytułów, wsparty na tradycji. Nazwiska i tytuły arystokratyczne wędrują wedle specjalnych praw: i tak prince des Laumes zaczął się tytułować od śmierci swego ojca duc de Guermantes, podczas gdy jego rodzony brat zowie się baron de Charlus itd."], "slug": "w-cieniu-zakwitajacych-dziewczat-tom-pierwszy", "xml_sha256": "ef1e30963eb916f68bd4886dd1d0beb9c4fa1e65c2206485cb48f28d35a86041"}
-{"date_pd": "2017", "id": "wolne_lektury_2371705", "pa": ["nie zauważyłbym chyba owego pochylenia podłogi --- być może podłoga nie była pochyła, lecz samo muzeum zbudowane zostało na zboczu wzgórza."], "pt": [], "slug": "wells-wehikul-czasu", "xml_sha256": "3b8a28b1811a635a74da6aebc347372ea5a2206de17f4d98f676b7e2ec544831"}
 {"date_pd": "2012", "id": "wolne_lektury_2371708", "pa": [], "pt": ["do Hiszpanii (...) sprawy rodzinne --- chodziło o siostrę pisarza uwiedzioną przyrzeczeniem małżeństwa przez niejakiego Clavijo. Beaumarchais opisał później w swoich Memoriałach całe zdarzenie i swoją interwencję w sposób tak żywy i dramatyczny, iż opis ten stał się natchnieniem Wolfganga Goethego (1749--1832) do jego tragedii Clavigo.", "Grimm --- tu: Fryderyk (1723--1807), niem. pisarz i krytyk literacki, piszący po francusku.", "Charles Collé (1709--1783) --- piosenkarz francuski.", "Trianon --- nazwa dwóch pałaców w Wersalu, tu mowa o Petit Trianon (Małym Trianon).", "Scribe (...) Sardou, Caillavet i Flers --- Eugéne Scribe (1791--1861): autor wielu komedii i librett operowych; Victorien Sardou (1831--1908): dramaturg francuski, autor m.in. Toski; Gaston Caillavet (1869--1915), Robert Flers (1872--1927): autorzy wspólnie pisanych komedii, jak np. wspomniany Król.", "Marivaux, Pierre (1688--1763) --- dramaturg francuski, którego komedie odznaczają się subtelną analizą psychologiczną i zdumiewającą lekkością stylu; pisał także powieści realistyczne jak np. Życie Marianny czy Kariera wieśniaka.", "Plautus (ur. ok. 254, zm. 184 p.n.e.) --- słynny komediopisarz rzymski, którego utwory jak np. Kupiec czy Żołnierz fanfaron wywarły duży wpływ na późniejszych dramaturgów z Molierem na czele.", "Sarcey, Francisque (1827--1899) --- francuski krytyk teatralny.", "Rabelaisowska figura sędziego --- postać komiczna z powieści Gargantua i Pantagruel Rabelais'go (1495--1553).", "ustęp apoteozujący szlachectwo, który sam [Beaumarche] z własnych Memoriałów cytuje w przedmowie --- ,,Nie iżby trzeba było zapominać (powiedział pewien poważny pisarz, i tym chętniej go cytuję, ile że jestem w tym jego zdania), nie iżby trzeba było zapominać, mówi, o tym, co się jest winnym wysokim stanowiskom; słuszna jest, przeciwnie, aby przywilej urodzenia najmniej ze wszystkich podawany był w wątpliwość. To darmo otrzymane dobrodziejstwo dziedzictwa, owoc czynów, cnót i przymiotów przodków, po których bierze się je w spadku, nie może zgoła ranić miłości własnej tych, którym go los odmówił. W monarchii, gdyby się usunęło pośrednie stopnie, byłoby za daleko od monarchy do poddanych; niebawem ujrzałoby się jedynie despotę i niewolników. Utrzymanie stopniowej drabinki, od pracującego na roli aż do potentata, leży w jednakim stopniu w interesie wszystkich stanów i jest może najsilniejszą podporą monarchii\".", "Grzegorz Dyndała --- postać z komedii Moliera pod tym samym tytułem; Dyndała, z pochodzenia chłop, poślubia szlachciankę, która go zdradza i ośmiesza przy każdej okazji. Powiedzenie Dyndały: ,,Ha, sam chciałeś, sam chciałeś, Grzegorzu!\", stało się we Francji przysłowiowe.", "pojawiły się w polskim przekładzie niemal bezpośrednio po ich ogłoszeniu drukiem w języku francuskim --- Cyrulik, sewilski, Warszawa 1780; Wesele Figara, Warszawa 1786.", "piosenka Cherubina --- przekład Edwarda Leszczyńskiego.", "w 1783 r. książę de Nassau przebywał w Hiszpanii, Paryżu i Wiedniu, a żona jego w Paryżu --- porównaj: D'Aragon, Un paladin au XVIII-e siècle. Le prince Charles de Nassau-Siegen, Paris 1893."], "slug": "wesele-figara", "xml_sha256": "5639b2bd69d7109e67e0c9ecde8ee6589e62940391df480461a3c9d210d2d1c1"}
 {"date_pd": "1981", "id": "wolne_lektury_2371714", "pa": ["głębokie poręczowe krzesło --- łac. cathedra. Guhl i Koner Das Leben der Griechen und Römer etc. Berlin, 1882, str. 568.", "owijająca głowę zasłona --- łac. suffibulum. Zawój westalki, patrz Guhl i Koner str. 722.", "w mętnem zwierciedle zmysłów ludzkich --- Zwierciadła u rzymian. Guhl i Kohner, str. 656--658. fig. 490. Metalowe ręczne i stojące, małe i wielkości człowieka dosięgające. Ze spodu złocone. Etruskie z Prenesty.", "za połowę obrotu wodnego zegara --- obrót wodnego zegara rzymian trwał minut dwadzieścia. Pliniusz młodszy Listy.", "uczta śmiesznego Trymalchiona --- Uczta Trimalchiona, w Satyrykonie, utworze Petroniusza, napisanym za panowania Nerona.", "dziewice Westy --- łac. Virgines Vestae.", "Przebaczenie dla skazańca, któremu na spotkanie wyszły dziewice Westy! Niech wolny odejdzie --- skazaniec, wiedziony na śmierć, jeżeli trafem spotkał westalkę, to według praw rzymskich, otrzymywał wolność. Guhl i Koner, str. 723.", "Gdy z szyi mi zdejmą tę złotą kulę --- do lat 17-u młodzi rzymianie nosili na szyi kulę złotą, na łańcuszku zawieszoną, jako znak wieku dziecinnego. Po skończeniu lat 17-u oblekano ich w t. zw. togę dojrzałości. O zwyczaju tym wiele wzmianek w poezyach Marcyalisa.", "dekada --- dni dziesięć. Miesiąc u rzymian dzielił się na trzy dekady.", "sophos! --- po grecku: brawo!", "latiklawa --- łac. laticlavia tunica, tunika bramowana purpurą, noszona tylko przez senatorów i patrycyuszów.", "Cieniu Owidyuszowy, natchnij mię (...) bez Owidyuszowego nauczania --- Owidyusz był twórcą dzieła p. t. Sztuka kochania (Ars amandi).", "edylowie rzymscy --- edylowie pełnili w Rzymie czynność dzisiejszéj policyi.", "tron --- łac. thronus. Guhl i Koner, str. 569.", "Gemonie --- łac. scalae gemoniae, wschody wykute w skale, z których trupy śmiercią karanych przestępców wrzucano do Tybru. Valer. Maximus, Dio Cassius.", "jakby na widok Meduzy w posąg się przemieniła --- podług mitów greckich, widok głowy Meduzy przemieniał człowieka w posąg. Owidyusz: Metamorfozy. Meduza, jedna z Gorgon, strasznych istot z wężami na głowie.", "jest nas sześć --- westalek było jednocześnie zawsze sześć. Tytus Liwiusz 1, 20.", "do źródła Egeryi (...) zaopatrywanie świątyni w wodę jest jej świętą powinnością --- świątynia Westy powinna była codziennie być skrapiana wodą, czerpaną ze źródła Egeryi, znajdującego się w lasku tegoż imienia, w pobliżu Rzymu. Guhl i Koner, str. 723.", "plac Łotrów --- łac. campus sceleratus, Guhl i Koner, str. 724; Lübker Reallexicon des classischen Alterthums, str. 517.", "arcykapłan --- łac. Pontifex Maximus.", "Usuńcie widoki trupów, oręża, więzów i wszystkiego, co nieczyste --- wzbronienie powyższych widoków stosowało się szczególnie do wielkiego ofiarnika, noszącego tytuł: Flamen dialis, w części jednak i do arcykapłana. Guhl i Koner, str. 720.", "kolegium kapłańskie --- łac. collegium pontificum. Członkowie kolegium kapłańskiego nosili tytuły: pullarii, victimarii, tibicines, calatores, etc. etc.", "przedmiot święty --- łac. res sacra.", "świętość publiczna (...) narodu --- łac. sacra popularia, albo sacra populi Romani."], "pt": [], "slug": "westalka", "xml_sha256": "ddc8ac92d22b0b0283f75dec3c2b53be460842bb07e298129cd81cac2a9b00a0"}
 {"date_pd": "2012", "id": "wolne_lektury_2371770", "pa": [], "pt": ["Tybota, pana Ossyńskiego... --- Thibaut d'Auxigny, biskup Orleański, który trzymał Villona w ciężkim więzieniu w Meung.", "Na cnego Kotra mistrza... --- mistrz Jan Cotard, prokurator, świeżo zmarły; w dalszym ciągu Villon poświęca mu wspomnienie, a nawet osobną balladę.", "Pikardzki --- Pikardami nazywano heretyków, którzy nie uznawali modłów za zmarłych [Pikardowie --- grupa religijna, głosząca bliski koniec świata, funkcjonująca w XIV-XV w. w Europie Zachodniej, a następnie w Czechach - Red.WL].", "Psalm: Deus laudem... --- Psalm 108: Deus laudem meam, etc. Siódmy werset, który służy Villonowi jako modlitwa za biskupa Orleańskiego, opiewa: ,,Niechaj dni jego będą skrócone do najmniejszej liczby, a biskupstwo jego niech przejdzie w inne ręce\" [przytoczony przez Boya werset jest dokładnym cytatem z Wulgaty, to jest łacińskiej wersji Biblii; ,,biskupstwo\" trzeba tam jednak rozumieć jako ,,władzę (kapłańską)\"; Red.WL].", "Y cnemu królu Ludwikowi... --- Ludwik XI, którego przejazdowi przez Meung Villon zawdzięczał uwolnienie.", "Nad Arystotem Awerresa... --- Averroes, szanowany w średniowieczu lekarz i filozof arabski z XII w., komentator Arystotelesa.", "Romans Róży --- Roman de la Rose, ulubiony poemat średniowiecza [poemat alegoryczny o miłości, rozpoczęty przez Guillaume'a de Lorris (ok. 1235), a dokończony przez Jeana de Meung; Red.WL].", "Za Alexandra króla... --- Aleksander Wielki.", "Walery --- Valerianus Maximus; ale Villon, cytując z pamięci, przypisuje temu pisarzowi anegdotkę, którą prawdopodobnie zaczerpnął gdzie indziej, w Salisburym lub Janie de Vignay.", "Żałuię czasu mey młodości... --- Villon, pisząc testament, miał lat trzydzieści!", "Różnie los sadza ludzkie syny --- Hiob 7, 6.", "iako Dawid prawi... --- Quaesivi eum et non est inventus locus eius (Ps. 36) [Tekst łac. tłumaczy się jako: Szukałem go, lecz nie znalazłem miejsca, gdzie by był --- Red.WL].", "Refren tej ballady: Mais ou sont les neiges d'antan? należy do najsłynniejszych wierszy w literaturze francuskiej [współczesne tłumaczenie brzmi raczej: Ach, gdzie są niegdysiejsze śniegi --- Red.WL].", "Podobnież, gdzie ta monarchini... --- królowa Nawarry, która, wedle legendy, topiła swoich kochanków [Małgorzata Burgundzka (1290--1315) przyłapana na cudzołóstwie i uwięziona, a następnie uduszona na rozkaz męża, Ludwika Kłótliwego --- Red.WL].", "Berta --- legendarna matka Karola Wielkiego.", "Żale piękney płatnerki --- Owa płatnerka była historyczną osobistością: niegdyś głośna piękność paryska, za czasu Villona już zgrzybiała staruszka, nasunęła mu widokiem swoim te strofy, jedne z najlepiej odlanych w całym jego dziele [płatnerz --- handlarz broni i rynsztunku wojennego; Red.WL].", "Y ty, szewczycho Wilhelmino... --- w owych czasach wiele kobiet lekkiego obyczaju uprawiało rzekomo rozmaite rzemiosła, pod których godłem były znane, a które służyły im jako płaszczyk do bezpiecznego i tym zyskowniejszego praktykowania nierządu.", "Fremin --- ów Fremin, którego Villon nazywa żartobliwie ,,swoim klerkiem\" (sekretarzem), spisuje jakoby testament za dyktandem mistrza.", "Sardana, xiążę niezbyt słabe... --- prawdopodobnie Sardanapal.", "Kasia --- Katarzyna de Vauselles, jedyna postać kobieca, która znaczy się głębszym śladem w poezji Villona. W jakich okolicznościach odbyło się to ,,bicie\" --- nie wiadomo; słowa ,,na goło\" pozwalałyby się domyślać, iż Villon może został z urzędu oćwiczony: kara, na jaką skazywano wówczas żaków, gdy mścili się za srogość damy za pomocą zelżywych piosenek.", "Legacik mały tam spisałem... --- pierwszy znany utwór Villona pt. Les Lais (Legaty), zwany też Małym Testamentem.", "Moro, Proweniec, Roben, Trygam --- właściciele gospód, którym prawdopodobnie Villon został dłużny.", "Spadek zacnego imć bękarta... --- Piotr Marchant, zwany bękartem de la Barre, jeden z dwunastu sierżantów więzienia królewskiego w Châtelet.", "Item, dobremu oćcu --- więcey! ---/ Mistrzu Wilhelmu Wilonowi... --- przybrany ojciec i opiekun poety, kanonik przy klasztorze św. Benedykta, który widocznie wielokrotnie ratował wychowanka z ,,opressyi\".", "Powieść o dyabelskiey bździnie --- zaginiony poemat Villona, który miał za treść tragikomiczną walkę o kamień nazwany Pet-du-Diable.", "Tabaryn --- jeden z uczestników kradzieży w kolegium Nawarskim, który gadulstwem swoim sprawił, iż rzecz wyszła na jaw, a później, na torturze wodnej, wydał wspólników. Stąd prawdopodobnie ironiczna aluzja do jego ,,prawdomówności\".", "Ythier Marchant --- młodzieniec z zamożnego mieszczaństwa, towarzysz zabaw młodości Villona. Rondo, jakie Villon mu zapisuje, daje pojęcie o smaku ówczesnej dwornej poezji, takiej jaką uprawiali najcelniej Alain Chartier i Karol Orleański.", "Mistrzowi Rogatemu Ianowi... --- Jean le Cornu, bogaty mieszczanin i sekretarz królewski, który prawdopodobnie okazał się w potrzebie mało uczynnym dla poety. W ogrodzie, który mu ironicznie ,,zapisuje\", Villon, zdaje się, zmuszony był przespać nieraz noc.", "Saint-Amant --- urzędnik Skarbu; w oryg. legat ten zawiera aluzje do godeł sklepowych, jak Biały koń itd.; rodzaj żartu bardzo Villonowi ulubiony i powtarzany często.", "Ragier --- znany opój; ,,Wielki kubek\" --- oberża pod tym godłem.", ",,pod Szyszką\" --- Pomme du Pin, głośna oberża współczesna.", "Item, co Marbof, abo zgoła/ Mikołaj Luwier, z tymi bieda --- aluzje do dwóch znanych mieszczan, nadających sobie szlacheckie tony.", "Turgis --- właściciel głośnej traktierni Pomme du Pin.", "To iest, strażnicy, myślę, piesi... --- Villona obchodzili tylko strażnicy piesi, nie konni, ponieważ ci tylko odbywali ronty w mieście [ront --- patrol; Red.WL].", "Mistrz Jan Pulieński --- Jan de Poulieu, doktor uniwersytetu, który zwalczał zakony żebrzące i później musiał odwołać swoje zarzuty. Za czasu Villona istniała bardzo silna opozycja świeckiego kleru przeciw groźnej konkurencji zakonów żebrzących.", "Mistrz Jan Mehuński --- Jan de Meung, autor drugiej części Romansu Róży (1277).", "Zasię Matyska z Orleanu... --- Macé d'Orléans był to sędzia prowincjonalny, którego Villon, prawdopodobnie dla jego gadulstwa i drobiazgowości, przechrzcił na kobietę. Kobietom nierządnego życia nie wolno było, pod grzywną, nosić niektórych ozdób, np. pasków ze srebra.", "mistrzowi Lorensowi... --- mistrz Lorenz, promotor, który prowadził śledztwo w sprawie kradzieży nawarskiej, przy tym tęgi bibuła [bibuła --- pijak; Red.WL].", "mistrzowi Ianu Kotru... --- Jan Cotard, prokurator, również tęgi pijak, wówczas świeżo zmarły.", "Szpaczek młodszy --- syn bogatego bankiera, prawdopodobnie towarzysz uciech młodości Villona.", "biedne trzy sieroty --- trzej starzy lichwiarze, którym Villon poświęcił już aluzje w Legatach (Marcel, Gassouyn i Laurens).", "donatem --- gra słów na czasowniku dawać i tytule znanej gramatyki łacińskiej [łac. dono znaczy dawać; Red.WL].", "Credo --- gra słów na wyrazie Credo i kredycie [Credo --- modlitwa ,,Wierzę w Boga Ojca\", wyznanie wiary; Red.WL].", "Kupcie im ciastek... --- w oryg. gra słów na słowie flaon, ciastko, a także forma do bicia monety.", "Oto mi dziatki dobrey woley! --- ,,enfants de lieu de bien\", wyrażenie, którym określano najgorszych hultai i oszustów.", "mym biednym klerykusiom --- dwaj starzy kanonicy przy kościele Notre-Dame, z którym to kościołem klasztor św. Benedykta był w zajadłej nieprzyjaźni.", "Na domu Giedryia Wilhelma... --- osobistość znana z niewypłacalności.", "By zacną bursę biedne żaczki... --- właśnie ci dwaj kanonicy rozdzielali miejsca w bursie, którymi ich Villon ironicznie obdarza.", "Nie śpią tak, iako owe szpaczki... --- prawdopodobnie właśnie rzecz miała się odwrotnie.", "Item, imć Kuldu Michałowi,/ Y Karlotowi imć Taranie --- dwaj bogaci mieszczanie paryscy. Żółte buty były wówczas cechą wysokiej elegancji.", "imć panu Grynieńskiemu... --- Filip Brunei de Grigny, człowiek zrujnowany i wiecznie goniący za pieniędzmi.", "Bassanier --- rejent i pisarz w trybunale. Pierniczkami nazywano, w języku ówczesnej palestry, kubany dawane urzędnikom [,,kuban\" oznacza łapówkę; Red.WL].", "Zacnemu panu prefektowi... --- Robert d'Estouteville, ożeniony z córką poprzedniego prefekta Ambrozją Delore; swego czasu Villon, zaszczycany łaskami prefekta, napisał dlań balladę, którą tutaj pomieścił.", "W turnieiu króla Reneusa... --- René, książę Prowansji i król Sycylii.", "Ni Franciszkowi, bratu iego... --- prawdopodobnie obaj bracia okazali się nieużytymi dla poety; prócz tego Franciszek oddał mu jakąś złą przysługę wobec młodego arcybiskupa z Bourges.", "Ieśli na rozdział w Taiawencie... --- głośna ówczesna książka kucharska.", "Przeciwrzeczenia Fran-Gontirowe --- biskup z Meaux, Filip de Vitry, napisał głośną współcześnie sielankę, której bohaterem jest ubogi drwal Franc-Gontier i żona jego Helena, a treścią apoteoza szczęścia w ubóstwie. Sielanka ta stworzyła mnóstwo naśladownictw i cały styl: opływający w dostatki panowie, nawet René, książę Prowansji i król Neapolu, rozczulali się nad rozkoszami ubóstwa, co widocznie podrażniło Villona, który z gorzkiego doświadczenia wiedział, co o tym szczęściu myśleć. Ballada, która następuje, jest odpowiedzią na te niewczesne zachwyty; a poświęcona jest mistrzowi Couraud, który był pełnomocnikiem sielankowego króla René w Paryżu. Być może, iż Villon szukał za jego pośrednictwem poparcia u księcia Prowansji i, zamiast spodziewanej pomocy, otrzymał tylko zbawienne rady. Do tego odnosiłaby się też wzmianka o ,,tyranie\".", "Pani de Bruier --- pani de Bruyeres, pobożna wdowa paryska, trudniąca się propagandą cnoty.", "Z Małego Mostu... --- murowany most na Sekwanie, cały zabudowany i będący najbardziej ożywionym punktem handlowym Paryża.", "Makrob --- pisarz z V w., autor Saturnalii, w których ks. VII, p. 7, mieści się rozprawa o tym, dlaczego natura kobiet gorętsza jest niż mężczyzn: ,,Quid plura? Nonne videmus mulieres, quando nimium frigus est, mediocri veste contentas, nec ita operimentis plurimis involutas, ut viri solent, scilicet naturali calore, contra frigus, quod aer ingerit, repugnante? [Cóż więcej? Czyż nie widzieliście kobiet, które, chociaż jest zimno, zadowalają się skromną odzieżą, jak to czynią mężczyźni, jak gdyby zwalczały chłód, jaki przynosi powietrze, naturalnym ciepłem? --- tłum. Red.WL]", "Ten klasztor... --- klasztor żeński na Montmartre, znany ze swobody obyczajów i łatwego dostępu dla mężczyzn.", "Brelar Bigod! --- przekleństwo w zepsutej niemczyźnie: Verloren, bey Gott!", "Wyiąwszy ieno kaźń Mehyńską... --- więzienie, które Villon dopiero co opuścił.", "Gładkiemu Noelowi... --- Noël Jolis, który, jak się zdaje, odegrał dwuznaczną rolę w czasie amorów poety dla Katarzyny de Vausselles i ich smutnego końca.", "Z ręki Henryka... --- Henry Cousin, kat miasta Paryża.", "Żebrzącym Braciom... --- nie pierwsza to aluzja w Testamencie, wymierzona przeciw zakonom żebrzącym: Villon był w tym wiernym echem poglądów i antypatii uniwersyteckiego duchowieństwa paryskiego.", "Item, moiemu balwierzowi, Tuż wpodle Angla herborysty, Imieniem Mrozik Kolinowi, Sopelek lodu... --- dość prostoduszna igraszka słów z nazwiska Colin Galerne, znanego balwierza paryskiego.", "Do Montpippeau y do Ruclu... --- miejscowości opodal Paryża.", "Koleń Kaieński --- Colin de Cayeux, niepoprawny włamywacz, należący do jednej szajki z Villonem, powieszony z wyroku sądowego.", "Dydona --- królowa Kartaginy, bohaterka Eneidy, uchodziła w średnich wiekach za najwyższe wcielenie powabów niewieścich.", "Co wnet po śmierci ludzi czerni... --- aluzja do szubienicy [i/lub do zarazy nazywanej Czarną Śmiercią; Red.WL].", "Osobno grześć u Niewiniątek --- ociemniali miasta Paryża mieli przywilej kwestowania w dnie świąteczne na cmentarzu Niewiniątek.", "Kiedy na one patrzę głowy... --- cmentarz ten służył Paryżanom za miejsce przechadzek i zabaw. Często zachodziła potrzeba rozkopywania grobów dla zyskania miejsca, przy czym zesypywano kości na jedną kupę, tak iż w istocie przechadzająca się publiczność ocierała się o ,,kopice\" czaszek ludzkich, wszelakiego stanu i kondycji.", "Chciwości wrogi y zakały... --- wyrażenie oczywiście ironiczne, tak jak i dalsze zestawienie św. Dominika z Panem Bogiem co do potęgi.", "mistrzowi Lomerowi... --- mistrz Lomer d'Airaines, zakonnik; warunki stawiane przez testatora czynią prawo mistrza Lomera dość problematycznym zwłaszcza wobec pojęć Villona o interesowności kobiecej.", "rycerz Ogier Wielki --- legendarny rycerz Ogier le Danois, słynny ze swej dzielności w turniejach miłosnych.", "Z Szartierowego kwartą mleczka... --- Alain Chartier, poeta współczesny, autor czułych pieśni miłosnych; w oryg. gra słów lait, mleko i lais, piosenka.", "Żemsowi Iakobowi... --- mistrz Jakob James, skąpiec i właściciel licznych domów, nie cieszących się zbyt dobrą reputacją.", "dla imci Seneszala... --- być może Piotr de Brezé, wielki pan a zarazem poeta, wielki seneszal Normandii, uwięziony przez Ludwika XI w Loches.", "Ianowi Kale... --- Jean de Calais, notariusz Châtelet, w istocie piastujący funkcje interpretatora testamentów.", "Pogrzebią u Iadwigi świętey... --- klasztor stanowiący schronienie dla starych niezamężnych kobiet.", "By nie obciążać zbyt podłogi... --- kaplica tego klasztoru mieściła się na pierwszym piętrze.", "Srogim wyrokiem przepędzony... --- ten ,,srogi wyrok\", wydany ,,wbrew apelacji\", można odnosić do srogości damy serca Villona, którą już poprzednio czynił sprawczynią swojej tułaczki.", "W dzwon znaczny... --- wielki dzwon na wieży kościoła Notre-Dame.", "Legatu cne exekutory... --- egzekutorów [wykonawców; Red.WL] tych wyszczególnia Villon w kilku strofkach, które w przekładzie opuszczono.", "List do przyiacioł, w formie ballady --- list, pisany z więzienia w Meungs, do towarzyszy życia i zabaw w Paryżu.", "Weźrzyicie, ieśli wola, na sierotę! (...) W onem wygnaniu, kędy w żalu trawię --- bogate określenie na ,,obywateli świeżego powietrza\".", "Niech nikt z naszego nieszczęścia nie szydzi... --- wisielcy stanowili ulubiony temat popularnych żartów i konceptów.", "Stefan Garnier --- osobistość wielce wątpliwej moralności, prawdopodobnie dobry znajomy Villona.", "Kapetowie --- królewski ród Kapetów wiódł się, wedle popularnej gadki, od rzeźnika.", "Rozumiesz te figielki?... --- aluzja do tortury wodnej, jakiej poddano Villona w śledztwie.", "Rozprawa serca y ciała Wilonowego w kształcie ballady --- tego rodzaju ,,rozprawy\" między ciałem a duszą istniały już i przedtem w literaturze średniowiecza.", "druga część --- autorem pierwszej, wcześniejszej o jakie pół wieku, jest Wilhelm de Lorris.", "Tradycję tę utrwalił poemacik p. t. Repues franches (Daremne biesiady), którego bohaterem jest Villon i którego autorstwo przypisywano nawet długo jemu samemu.", "napisał w złodziejskim narzeczu szajki szereg ballad o treści zaczerpniętej z życia włóczęgów.", "Katarzyna de Vausselles --- poszukiwania co do osoby Katarzyny de Vauselles pozostały daremne. Tyle wykryto, że istniała rodzina tego nazwiska i że mieszkała, za czasów Villona, w pobliżu klasztoru św. Benedykta, który służył za dom poecie.", "Pośmiertne losy spuścizny Villona znaczą się nader interesującą linią... --- Warto zauważyć, iż Montaigne, wśród swojej, obejmującej tyle tematów gawędy, nie zdradza ani słowem znajomości Villona; Rabelais znał go niewątpliwie i wspomina w wielu miejscach, lecz raczej w duchu pospolitej legendy, już utworzonej o pisarzu, bez głębszego wniknięcia w jego treść.", "Teofil Gautier --- znakomity krytyk, który nie przeoczył chyba ani jednego momentu literatury francuskiej, Sainte-Beuve, poświęca, w r. 1859, Villonowi jeden ze swych poniedziałków, z mniejszym jednak niż zazwyczaj odczuciem i wżyciem się w indywidualność pisarza.", "wilonologia --- twórcą nowoczesnej ,,Wilonologii\" jest August Longnon, który opracował pierwsze krytyczne wydanie dzieł poety (Paris, A. Lemerre, 1892), jak również wykrył cały szereg danych biograficznych; dalej autor licznych prac tyczących poety i znakomity jego monografista (Les grands écrivains français, 1901), Gaston Paris; dalej niestrudzony badacz Marceli Schwob; wreszcie towarzysz jego prac, Piotr Champion, który dotychczasową wiedzę o Villonie zamknął w pomnikowym wydawnictwie: François Villon, sa vie et son temps (Paris, II. Champion, 1913). Na podstawie wszystkich tych prac oparłem moje przypisy. W ostatnich latach ukazały się duże urojone biografie Villona, stylizowane dzisiejszą modą w formie powieści: Pierre d'Allheim, La Passion de maitre François Villon (1924) i Francis Carco Le roman de François Villon (1926)."], "slug": "wielki-testament", "xml_sha256": "beb00a6cec7d23fc1cdd73a336d081fcbffa07f6f7fd3d458f2ccdd2f62b5ab2"}

data/wolne_lektury/wolne_lektury.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:68033c371034eee9311f3838a6e4d6da07691870392b8ef8bd9037289269fadb
-size 114922621
+oid sha256:d223859e1f81f7dceb9269856077e4c265dda7d061a05b81576d85916f81606b
+size 114516870

data/wolne_lektury/wolne_lektury.stats.json

@@ -1,26 +1,27 @@
 {
   "read": 6141,
-  "kept": 6076,
+  "kept": 6074,
   "drop_short": 0,
   "drop_lang": 0,
   "drop_dup": 0,
   "drop_ocr": 0,
-  "chars": 234671405,
-  "tokens": 92151055,
+  "chars": 233812331,
+  "tokens": 91823395,
   "licenses": {
     "CC-BY-SA-3.0": 5,
     "CC-BY-SA-3.0-PL": 34,
     "CC-BY-SA-4.0": 11,
     "LAL-1.3": 1467,
-    "public-domain": 4559
+    "public-domain": 4557
   },
-  "authors_with_value": 6039,
+  "authors_with_value": 6037,
   "license": "per-record Wolne Lektury licence (public domain, LAL 1.3, CC BY-SA 3.0 PL/3.0/4.0)",
   "stats_recomputed_from_parquet": true,
   "drop_eltec_dup": 63,
   "input_sha256": "5833c25db296f2479b3c09745ebedd550b2425dd6ebaf021f0095e684a0221a1",
   "eltec_sha256": "ba40061741dff11165149f6c2cf3d5526e1b6c759e7ce1e03ccc164d58838325",
   "drop_rights": 2,
-  "notes_rows": 460,
-  "notes_sha256": "07fe95a8265f6b40512305cf39b4960fbd16ec2ba9c4c68f7ce92b681d5a51a1"
+  "notes_rows": 459,
+  "notes_sha256": "da18d6a7eebf88f63efbe1e082d5e2cdc387bf9bbf808cc9d0a6411dbf14d8e1",
+  "drop_not_free": 2
 }

src/attribution_common.py

@@ -93,13 +93,23 @@ def text_checks(table: pa.Table) -> dict:
     }
 
 
+# A bracket closing the given names: the real name behind a pseudonym, as in 'M-ski, Adam [Zofia Trzeszczkowska]'.
+TRAILING_BRACKET = re.compile(r"(.*?)\s*(\[[^\[\]]*\])")
+
+
 def natural_order(name: str) -> str:
-    """'Sienkiewicz, Henryk' -> 'Henryk Sienkiewicz'. A name without exactly one comma, or with a
-    parenthesis (a qualifier such as '(ojciec)'), stays as the source gives it."""
+    """'Sienkiewicz, Henryk' -> 'Henryk Sienkiewicz', with runs of white space made one space. A bracket after the given
+    names stays last ('M-ski, Adam [Zofia Trzeszczkowska]' -> 'Adam M-ski [Zofia Trzeszczkowska]'). A name without
+    exactly one comma, with a parenthesis (a qualifier such as '(ojciec)') or with only a bracket after the comma stays
+    as the source gives it."""
+    name = " ".join(name.split())
     family, comma, given = (part.strip() for part in name.partition(","))
     if not comma or not family or not given or "," in given or "(" in name:
-        return " ".join(name.split())
-    return f"{given} {family}"
+        return name
+    bracket = TRAILING_BRACKET.fullmatch(given)
+    if not bracket:
+        return f"{given} {family}"
+    return f"{bracket[1]} {family} {bracket[2]}" if bracket[1] else name
 
 
 def write_sidecar(path: Path, records: list[dict]) -> None:

src/build_wolne_lektury_attribution.py

@@ -8,13 +8,15 @@ src/wolne_lektury_licences.py as <cache>/hdr/<slug>.rdf) and writes
   data/wolne_lektury/wolne_lektury.attribution.jsonl   one line per Parquet row, same order, sorted keys:
       id, slug, work_url, title, author, authors, translators, credits, source_edition, source_url,
       publisher, credit_notes, credit_line, license, license_url, wl_rights, wl_license_url, wl_match,
-      wl_date_pd, modification_note, modified, text_sha1
+      wl_date_pd, modification_note, modified, text_sha1, and on a few rows metadata_note
   data/wolne_lektury/wolne_lektury.parquet             `author` filled from the sidecar; nothing else changes
   data/wolne_lektury/wolne_lektury.stats.json          `authors_with_value`
   data/wolne_lektury/wolne_lektury.license-evidence.json  counts, normalisation steps and check results
 
 Names are WL's own credits turned from "Family, Given" into "Given Family"; WL's pseudonyms and cryptonyms
-stay as WL gives them. WL does not say which edition a name or title follows, so the first-edition form
+stay as WL gives them. The check of WL's credits against the National Library's catalogue (BN; `bn_check` in the
+evidence file) changed the translators of three rows (TRANSLATOR_FIXES) and annotated a fourth (SOURCE_EDITION_NOTES);
+these rows say so in `metadata_note`. WL does not say which edition a name or title follows, so the first-edition form
 that the maintainer asked for is NOT established here. `author` is the authors, and "Autorzy oryginału: A;
 Tłumaczenie: T" for a translation, the form the global_voices sidecar uses. The placeholders WL puts in
 dc:creator for an unknown or collective author ("Autor nieznany", "Praca zbiorowa") are not credited in
@@ -52,15 +54,15 @@ from attribution_common import (BUILD_DYNAWORD_STEPS, natural_order, steps_with_
 from sejm_api_common import check_pin, sha256_file  # noqa: E402
 from shingle_containment import strip_wl_footer  # noqa: E402
 from wolne_lektury_licences import (CC_BY_SA_3, CC_BY_SA_3_PL, CC_BY_SA_4, DATA, LABELS, LAL,  # noqa: E402
-                                    LICENCES, PUBLIC_DOMAIN, SOURCE, classify, read_licences)
+                                    LICENCES, NOT_FREE, PUBLIC_DOMAIN, SOURCE, classify, read_licences)
 from wolne_lektury_notes import (HEADINGS, NOTES, NOTES_SHA256, OWNER, TEXT_GLOSSES, read_notes,  # noqa: E402
                                  restoration)
 
 SIDECAR = f"{SOURCE}.attribution.jsonl"
 EVIDENCE = f"{SOURCE}.license-evidence.json"
-INPUT_SHA256 = "6b5429ed5dcf8e9bf9404f856ebfcc09bf5f178fe79e8aeffeb1db468dd0501f"  # src/clean_wolne_lektury.py's output
-TEXT_DIGEST = "a0d669c4c83fc9c624bc48eb610773d7d363f152523630a196990184527df8ab"  # text_digest of both Parquets
-OUTPUT_SHA256 = "68033c371034eee9311f3838a6e4d6da07691870392b8ef8bd9037289269fadb"  # the Parquet this script writes; test_attribution_release.py checks it
+INPUT_SHA256 = "cae78e0ea34138934839225c39bbb895c151a22a4be0f5b11ddd29d56df2d568"  # src/clean_wolne_lektury.py's output
+TEXT_DIGEST = "9fcc2c3e8e5a7bb3b80bc61fcb086a7148613144bde1adfc067bb6e92afa2b23"  # text_digest of both Parquets
+OUTPUT_SHA256 = "d223859e1f81f7dceb9269856077e4c265dda7d061a05b81576d85916f81606b"  # the Parquet this script writes; test_attribution_release.py checks it
 DC = "{http://purl.org/dc/elements/1.1/}"
 CREDIT_LINE = "Książka pochodzi z serwisu Wolne Lektury"
 # The full URL of each licence version, as named in WL's rights text. LAL is WL's own link (http).
@@ -141,6 +143,80 @@ EDITION_REVIEWED = {
     "wl_date_pd_wrong": {"wolne_lektury_2374660": "WL's date.pd 2012 is wrong (Konopnicka died 1910); the row is "
                                                   "public domain either way."},
 }
+# The check of WL's author and translator credits against the National Library's catalogue (BN, data.bn.org.pl), and
+# what the maintainer decided on it (polish-dynaword-v2#109). The rows it showed are not free are dropped upstream
+# (NOT_FREE in src/wolne_lektury_licences.py); bn_check() puts all of it in the evidence file.
+BN_CHECK = {
+    "checked_at": "2026-10-20",
+    "decided_at": "2026-10-20",
+    "source": "the National Library's catalogue, https://data.bn.org.pl (read only; WL's own e-book records left out "
+              "as circular)",
+    "method": "each row's BN records found by the author's surname and the title of the work and of its print "
+              "edition; our author and translators compared with BN's fields 100, 245$c and 700, and for public-domain "
+              "rows BN's death years with 1955 and with WL's date.pd. Every row flagged for an author or translator "
+              "was read by hand, the other flags in samples.",
+    "reviewed_by_hand": {
+        "in_full": {"translator_not_ours": 112, "translator_missing_in_bn": 53, "author_mismatch": 31,
+                    "death_year_after_1955": 1},
+        "in_samples": {"no_record": "50 of 503", "pseudonym": "50 of 76", "other": "48 of 120"},
+        "labels": {"real_error": 41, "catalogue_noise": 192, "pseudonym_ok": 36, "cannot_tell": 76},
+        "unit": "a row and flag; a row can have several flags",
+        "scope": "the check covered the 6,076 Wolne Lektury rows of the time (this table's and the 2 in NOT_FREE) and "
+                 "the 100 eltec_pol novels; 7 of the row-flags read are on eltec_pol rows (5 pseudonym_ok, 1 "
+                 "catalogue_noise, 1 cannot_tell)",
+    },
+    "public_domain_rows_with_a_bn_death_after_1955": "one, \"Wola mocy\", dropped (dropped_as_not_free)",
+    "open": "the cannot_tell rows not named in this section are left as Wolne Lektury gives them; they are listed in "
+            "polish-dynaword-v2#109",
+}
+# Rows whose translators in WL's header are wrong (maintainer decision 2026-10-20): WL's translators after
+# natural_order, ours, and the evidence. The build fails if WL's header stops giving the first list.
+TRANSLATOR_FIXES = {
+    "wolne_lektury_2374887": (["Leon Urlich"], ["Leon Ulrich"],
+                              "Wolne Lektury's own rights line reads \"Leon Ulrich zm. 1885\"; BN record 1504075 "
+                              "(Kraków: Gebethner, 1895), \"przekład J. Ulricha\", \"Ulrich, Leon (1811-1885)\"."),
+    "wolne_lektury_2373112": (["Szymon Dickstein"], ["Szymon Dickstein", "Józef Nusbaum"],
+                              "Wolne Lektury's print edition reads \"tłum. Szymon Dickstein i Józef Nusbaum\"; BN "
+                              "record 2114549 (Warszawa 1885), \"przeł. Szymon Dickstein i Józef Nusbaum\", "
+                              "\"Nusbaum-Hilarowicz, Józef (1859-1917)\"."),
+    "wolne_lektury_2372990": ([], ["Aniela Zagórska"],
+                              "Wolne Lektury's print edition reads \"tłum. Aniela Zagórska\" and its rights line "
+                              "\"Aniela Zagórska zm. 1943\"; BN record 1902875 (Warszawa: PIW, 1957), \"przeł. Aniela "
+                              "Zagórska\", \"Zagórska, Aniela (1881-1943)\"."),
+}
+# Rows whose print edition in WL's header (dc:source) is not the book of the text; the sidecar keeps WL's value.
+SOURCE_EDITION_NOTES = {
+    "wolne_lektury_2374125": "Wolne Lektury's print edition (`source_edition`) is Hebbel's play \"Judyta\" in Kazimierz "
+                             "Kaszewski's translation (BN record 1454550, Warszawa 1881), not this book, Hebbel's "
+                             "diaries in Karol Irzykowski's translation (Wolne Lektury's rights line: \"Karol Irzykowski "
+                             "zm. 1944\"). Which print the text follows is not established.",
+}
+# Montaigne's "Próby": Wolne Lektury credits Edmund Cięglewicz note by note for the translated Latin and Greek
+# quotations; BN credits Tadeusz Sinko (d. 1966) and Cięglewicz jointly. Kept (maintainer decision 2026-10-20).
+MONTAIGNE = {"wolne_lektury_2376515": "Księga pierwsza", "wolne_lektury_2376514": "Księga druga",
+             "wolne_lektury_2376516": "Księga trzecia"}
+MONTAIGNE_CREDIT = "tłum. Edmund Cięglewicz"
+MONTAIGNE_FINDING = (
+    "BN record 2293540 (Próby, Ks. 1, Warszawa: PIW, 1957, the edition Wolne Lektury names for all three rows) reads "
+    "\"[cytaty łac. i grec. przeł. Tadeusz Sinko i Edmund Cięglewicz]\", with \"Cięglewicz, Edmund (1862-1928)\" and "
+    "\"Sinko, Tadeusz (1877-1966)\". Wolne Lektury's print edition for Księga pierwsza reads \"cytaty łac. tłum. Edmund "
+    "Cięgielewicz\", its translator's notes end the citation of each quotation they translate with \"tłum. Edmund "
+    "Cięglewicz\", and none names Sinko. The notes of the other two books give the source of a quotation and seldom a "
+    "translation. Kept as Wolne Lektury credits them (maintainer decision 2026-10-20); which quotations BN's joint "
+    "credit covers is not established.")
+# What the check found and the maintainer left as Wolne Lektury wrote it (2026-10-20): none changes a row's status.
+WL_RIGHTS_DEATH_YEARS = {  # text in WL's rights line: the death year BN gives
+    "Jerzy Żuławski zm. 1914": 1915, "Bolesław Leśmian zm. 1938": 1937, "Jan Kiliński zm. 1760": 1819,
+    "Władysław Witwicki zm. 1947": 1948, "Wacław Rolicz-Lieder zm. 1940": 1912,
+}
+WL_DATE_PD_EARLY = {  # WL's date.pd is earlier than BN's death years give; every row is free today
+    "wolne_lektury_2372974": "the translator Franciszek Mirandola died 1930 (BN): free from 2001, not 1903",
+    "wolne_lektury_2374125": "the translator Karol Irzykowski died 1944 (BN): free from 2015, not 1981",
+    "wolne_lektury_2374971": "the translator Antoni Lange died 1929 (BN): free from 2000, not 1938",
+    "wolne_lektury_2375897": "the translator Cyprian Bazylik died 1600 (BN): free from 1671, not 1643",
+    "wolne_lektury_2376199": "the author Arthur Conan Doyle died 1930 (BN): free from 2001, not 1994",
+    "wolne_lektury_2376940": "the translator Maria Cumft-Pieńkowska died 1920 (BN): free from 1991, not 1971",
+}
 # The steps between SpeakLeash's file and the released text, from the build code. Only the first two and the
 # notes step can change a text; every other step keeps or drops whole documents.
 NORMALISATION_STEPS = [
@@ -153,6 +229,8 @@ NORMALISATION_STEPS = [
      "eltec_pol novel (5-word-shingle containment of at least 0.9)", "stat": "drop_eltec_dup"},
     {"stage": "src/clean_wolne_lektury.py", "changes_text": "no", "step": "drop a whole document for which Wolne "
      "Lektury gives no usable licence", "stat": "drop_rights"},
+    {"stage": "src/clean_wolne_lektury.py", "changes_text": "no", "step": "drop a whole document that Wolne Lektury "
+     "marks public domain although its translation is not free (bn_check below)", "stat": "drop_not_free"},
     {"stage": "src/clean_wolne_lektury.py", "changes_text": "no", "step": "set `license` from the document's own "
      "Wolne Lektury book"},
     {"stage": "src/clean_wolne_lektury.py", "changes_text": "yes, for the rows whose book has them",
@@ -249,6 +327,17 @@ def make_record(row_id: str, text: str, licence: dict, header: dict, kinds: list
         raise SystemExit(f"{row_id}: the header's work URL {header['work_url']!r} is not book {slugs[0]}")
     authors = [natural_order(a) for a in header["creators"]]
     translators = [natural_order(t) for t in header["contributors"].get("translator", [])]
+    notes = []
+    if row_id in TRANSLATOR_FIXES:
+        wl, fixed, evidence = TRANSLATOR_FIXES[row_id]
+        if translators != wl:
+            raise SystemExit(f"{row_id}: WL's header gives the translators {translators}, not {wl}; review "
+                             "TRANSLATOR_FIXES")
+        given = "; ".join(f"\"{name}\"" for name in wl) or "none"
+        notes.append(f"`translators` corrected from Wolne Lektury's header, which gives {given}. {evidence}")
+        translators = fixed
+    if row_id in SOURCE_EDITION_NOTES:
+        notes.append(SOURCE_EDITION_NOTES[row_id])
     credits = {role: [natural_order(n) for n in names] for role, names in header["contributors"].items()
                if role != "translator" and names}
     return {
@@ -260,6 +349,7 @@ def make_record(row_id: str, text: str, licence: dict, header: dict, kinds: list
         "id": row_id,
         "license": label,
         "license_url": LICENCE_URLS[label],
+        **({"metadata_note": " ".join(notes)} if notes else {}),
         "modification_note": modification_note(kinds, glossed, text_glosses),
         "modified": True,
         "publisher": header["publisher"],
@@ -402,6 +492,53 @@ def edition_reviewed(records: list[dict], headings: dict) -> dict:
     }
 
 
+def bn_check(records: list[dict], chosen: dict[str, dict]) -> dict:
+    """The check against BN's catalogue (BN_CHECK) and what came of it. The ids and counts are computed; the build
+    fails if a row it names is not in the table or a dropped one is."""
+    ids = {r["id"] for r in records}
+    named = TRANSLATOR_FIXES.keys() | SOURCE_EDITION_NOTES.keys() | MONTAIGNE.keys() | WL_DATE_PD_EARLY.keys()
+    if named - ids or NOT_FREE.keys() & ids:
+        raise SystemExit(f"bn_check names rows that are not in the table: {sorted(named - ids)}, or a NOT_FREE row is "
+                         f"kept: {sorted(NOT_FREE.keys() & ids)}")
+    death_years = {text: {"bn_death_year": year, "ids": [r["id"] for r in records if text in r["wl_rights"]]}
+                   for text, year in WL_RIGHTS_DEATH_YEARS.items()}
+    if not all(found["ids"] for found in death_years.values()):
+        raise SystemExit("a WL_RIGHTS_DEATH_YEARS text is in no row's rights line")
+    pseudonym = "Adam M-ski [Zofia Trzeszczkowska]"
+
+    def montaigne(row_id: str) -> dict:
+        notes = chosen.get(row_id, {}).get("pt", [])
+        credited = [n for n in notes if MONTAIGNE_CREDIT in n]
+        greek = [n for n in credited if "(gr.) ---" in n]
+        return {"book": MONTAIGNE[row_id], "translator_notes": len(notes), "crediting_cieglewicz": len(credited),
+                "of_them_greek": len(greek), "greek_characters": sum(map(len, greek)),
+                "naming_sinko": sum("Sinko" in n for n in notes)}
+
+    return {
+        **BN_CHECK,
+        "dropped_as_not_free": NOT_FREE,
+        "translators_fixed": {row_id: {"wolne_lektury": wl, "ours": fixed, "evidence": evidence}
+                              for row_id, (wl, fixed, evidence) in TRANSLATOR_FIXES.items()},
+        "source_edition_notes": SOURCE_EDITION_NOTES,
+        "pseudonym_with_real_name": {
+            "translator": pseudonym, "rows": sum(pseudonym in r["translators"] for r in records),
+            "meaning": "Zofia Trzeszczkowska wrote as Adam M-ski. Most of WL's headers give \"M-ski, Adam [Zofia "
+                       "Trzeszczkowska]\"; natural_order() keeps the bracket after the name."},
+        "joint_credit_kept": {
+            "finding": MONTAIGNE_FINDING,
+            "counted": f"translator's notes restored at the end of the text; crediting_cieglewicz: those that contain "
+                       f"\"{MONTAIGNE_CREDIT}\"; of_them_greek: those of them whose quotation WL marks \"(gr.)\"",
+            "rows": {row_id: montaigne(row_id) for row_id in MONTAIGNE},
+        },
+        "kept_as_wolne_lektury_wrote": {
+            "meaning": "the check found these and the maintainer left them as Wolne Lektury wrote them (2026-10-20); "
+                       "none changes whether a row is free",
+            "death_year_in_wl_rights": death_years,
+            "wl_date_pd_early": WL_DATE_PD_EARLY,
+        },
+    }
+
+
 def wl_commentary(records: list[dict], headings: dict) -> dict:
     """The rows with a `Komentarz` heading: all are WL's own teaching material, with its commentary in the text."""
     ids = set(headings.get("komentarz", {}).get("ids", []))
@@ -477,12 +614,14 @@ def build_evidence(table, records: list[dict], stats: dict, chosen: dict[str, di
         "notes_and_prefaces": {
             "scan_of_every_row": "text_checks above",
             "hand_sample": {"rows": NOTES_SAMPLE_SIZE, "draw": f"random.Random({NOTES_SAMPLE_SEED}).sample(ids, "
-                            f"{NOTES_SAMPLE_SIZE})", "read": f"the first and last {NOTES_SAMPLE_CHARS} characters "
+                            f"{NOTES_SAMPLE_SIZE}), ids: the 6,076 rows of the time in id order (this table's and the "
+                            "2 in NOT_FREE, neither drawn)", "read": f"the first and last {NOTES_SAMPLE_CHARS} characters "
                             "of each text", "result": NOTES_SAMPLE_RESULT},
             "wl_commentary": wl_commentary(records, checks["section_headings"]),
         },
         "footnote_check": footnote_check(chosen, left_out, stripped),
         "wl_glosses_in_main_text": text_glosses_check(),
+        "bn_check": bn_check(records, chosen),
     }
 
 

src/clean_wolne_lektury.py

@@ -21,7 +21,9 @@ Both groups are written to wolne_lektury.eltec_pairs.tsv.
 The `license` of each kept row comes from wolne_lektury.licences.tsv, which
 src/wolne_lektury_licences.py writes from Wolne Lektury's metadata; the build stops unless that file
 labels exactly the kept rows. The rows in that module's UNLICENSED, whose metadata gives no usable
-licence, are dropped first (stats `drop_rights`).
+licence, are dropped first (stats `drop_rights`), then the rows in its NOT_FREE, which Wolne Lektury marks public
+domain although the check against the National Library's catalogue showed their translation is not free (stats
+`drop_not_free`).
 
 The notes come from the committed wolne_lektury.notes.jsonl (pinned by SHA-256; see src/wolne_lektury_notes.py for the
 format, what is left out, and how to rebuild it from Wolne Lektury's book XML). They are added after the dedup, so the
@@ -59,7 +61,7 @@ from attribution_common import text_digest
 from sejm_api_common import check_pin
 from shingle_containment import best_containment, strip_wl_footer
 from sources import SOURCES
-from wolne_lektury_licences import LABELS, LICENCES, UNLICENSED, read_licences
+from wolne_lektury_licences import LABELS, LICENCES, NOT_FREE, UNLICENSED, read_licences
 from wolne_lektury_notes import (NOTES_SHA256, TEXT_DIGEST_BEFORE, TEXT_DIGEST_WITHOUT_GLOSSES, add_notes, read_notes,
                                  restoration, strip_text_glosses)
 
@@ -96,12 +98,12 @@ def relabel(table: pa.Table, licences_path: Path) -> pa.Table:
                             pa.array([labels[i] for i in ids], pa.string()))
 
 
-def drop_unlicensed(table: pa.Table) -> pa.Table:
-    """`table` without the UNLICENSED rows, every one of which must be in it."""
-    unlicensed = pc.is_in(table.column("id"), pa.array(sorted(UNLICENSED)))
-    if pc.sum(unlicensed).as_py() != len(UNLICENSED):
-        raise SystemExit("an UNLICENSED id is not among the deduplicated rows")
-    return table.filter(pc.invert(unlicensed))
+def drop_listed(table: pa.Table, ids, name: str) -> pa.Table:
+    """`table` without the rows in `ids` (the list called `name`), every one of which must be in it."""
+    listed = pc.is_in(table.column("id"), pa.array(sorted(ids)))
+    if pc.sum(listed).as_py() != len(ids):
+        raise SystemExit(f"an id in {name} is not among the deduplicated rows")
+    return table.filter(pc.invert(listed))
 
 
 def build(input_path: Path, eltec_path: Path, licences_path: Path, out_dir: Path) -> dict:
@@ -122,8 +124,9 @@ def build(input_path: Path, eltec_path: Path, licences_path: Path, out_dir: Path
                           "action": "drop" if score >= DROP_AT else "keep_partial",
                           "wl_title": wl_title(text)})
     deduped = table.filter(pa.array(keep))
-    licensed = drop_unlicensed(deduped)
-    labelled = relabel(licensed, licences_path)
+    licensed = drop_listed(deduped, UNLICENSED, "UNLICENSED")
+    free = drop_listed(licensed, NOT_FREE, "NOT_FREE")
+    labelled = relabel(free, licences_path)
     if text_digest(labelled) != TEXT_DIGEST_BEFORE:
         raise SystemExit("the text before the notes is not the text the notes step was written for")
     unglossed = strip_text_glosses(labelled)
@@ -139,6 +142,7 @@ def build(input_path: Path, eltec_path: Path, licences_path: Path, out_dir: Path
         kept=kept.num_rows,
         drop_eltec_dup=table.num_rows - deduped.num_rows,
         drop_rights=deduped.num_rows - licensed.num_rows,
+        drop_not_free=licensed.num_rows - free.num_rows,
         chars=pc.sum(pc.utf8_length(kept.column("text"))).as_py(),
         tokens=pc.sum(kept.column("token_count")).as_py(),
         licenses=dict(sorted(Counter(kept.column("license").to_pylist()).items())),
@@ -149,7 +153,7 @@ def build(input_path: Path, eltec_path: Path, licences_path: Path, out_dir: Path
         input_sha256=INPUT_SHA256,
         eltec_sha256=ELTEC_SHA256,
     )
-    drops = ("drop_short", "drop_lang", "drop_dup", "drop_ocr", "drop_eltec_dup", "drop_rights")
+    drops = ("drop_short", "drop_lang", "drop_dup", "drop_ocr", "drop_eltec_dup", "drop_rights", "drop_not_free")
     if stats["read"] - sum(stats[k] for k in drops) != stats["kept"]:
         raise SystemExit("stats drop counts do not add up to the kept rows")
 

src/sources.py

@@ -554,20 +554,20 @@ SOURCES = {
                      "these are Wolne Lektury's modernised versions of the print editions, not the "
                      "editions as first printed. The edition a text follows is the one in Wolne "
                      "Lektury's print-edition field, kept in the sidecar as `source_edition` (named "
-                     "for 6,032 of the 6,076 documents). Wolne Lektury's editorial notes are not "
+                     "for 6,030 of the 6,074 documents). Wolne Lektury's editorial notes are not "
                      "taken over (see the scope note).",
         "upstream": "https://wolnelektury.pl/api/books/",
         "provenance": "Pulled from SpeakLeash's public redistribution (`speakleash-ds-pub`, key "
                       "`wolne_lektury_corpus`) of Wolne Lektury's plain-text export. What SpeakLeash "
                       "did to the text is not documented, and who removed the export's footer is "
                       "not established. In this repository a text is changed in two ways: leading "
-                      "and trailing whitespace is trimmed, and the 460 texts whose Wolne Lektury "
+                      "and trailing whitespace is trimmed, and the 459 texts whose Wolne Lektury "
                       "book has author's or translator's notes get those notes as sections at the "
                       "end (\"Przypisy autora\", \"Przypisy tłumacza\"; Wolne Lektury's editorial "
                       "notes are not included; src/wolne_lektury_notes.py). No paragraph is "
                       "removed, and 63 whole "
-                      "documents were dropped as eltec_pol novels and 2 for want of a usable "
-                      "licence (src/clean_wolne_lektury.py). Wolne Lektury's metadata, fetched from "
+                      "documents were dropped as eltec_pol novels, 2 for want of a usable "
+                      "licence and 2 whose translation is not free (src/clean_wolne_lektury.py). Wolne Lektury's metadata, fetched from "
                       "its public API and book XML, fills `author` (\"Autorzy oryginału: A; "
                       "Tłumaczenie: T\" for a translation) and the sidecar; "
                       "src/build_wolne_lektury_attribution.py writes both and "
@@ -585,12 +585,26 @@ SOURCES = {
                       "translated by Paweł Kozioł, one marked only \"CC BY-SA\" with no version, the "
                       "other public domain for Marceli Motty's translation and \"CC BY-SA\" for "
                       "Kozioł's (polish-dynaword-v2#57). "
-                      "Flagged, nothing excluded (polish-dynaword-v2#109). `author` is Wolne Lektury's "
+                      "2 more are dropped after a check of Wolne Lektury's credits against the "
+                      "National Library's catalogue (polish-dynaword-v2#109): Wolne Lektury marks "
+                      "both public domain, but Nietzsche's \"Wola mocy\" is a joint translation by "
+                      "Konrad Drzewiecki and Stefan Frycz (d. 1960), and Wells's \"Wehikuł czasu\" "
+                      "follows a 1985 print whose translation an editor born in 1932 revised. The "
+                      "same check corrected the translators of 3 documents and found that the print "
+                      "edition named for a fourth (Hebbel's diaries) is another book; these 4 carry a "
+                      "`metadata_note` in the sidecar. In Montaigne's Próby (3 documents) Wolne "
+                      "Lektury's translator's notes credit the translated quotations to Edmund "
+                      "Cięglewicz, while the National Library's record of the print edition credits "
+                      "Tadeusz Sinko (d. 1966) and Cięglewicz jointly; which quotations each "
+                      "translated is not established. Wrong death years in the rights lines of 25 "
+                      "documents and early public-domain dates of 6 are kept as Wolne Lektury gives "
+                      "them; none changes whether a document is free. "
+                      "Otherwise flagged, nothing more excluded (polish-dynaword-v2#109). `author` is Wolne Lektury's "
                       "name form (pseudonyms and cryptonyms as Wolne Lektury gives them); the first-"
                       "edition form of a name is not established. 37 documents have no author "
                       "because Wolne Lektury gives only \"Autor nieznany\" or \"Praca zbiorowa\"; "
                       "Wolne Lektury has no field for the original title of a translation. Of the "
-                      "public-domain documents, 1,311 name a print edition published after 1955, "
+                      "public-domain documents, 1,310 name a print edition published after 1955, "
                       "58 a critical edition and 229 an introduction, notes or commentary; the "
                       "editors of those editions and their death years are not established. 239 "
                       "documents keep period spelling by our heuristic count, so the modernisation "
@@ -600,8 +614,8 @@ SOURCES = {
                       "print edition (a publisher's address, a list of editions, an introduction) "
                       "and none carried Wolne Lektury's credits or notes. The export had no "
                       "footnotes. From Wolne Lektury's book XML (read 2026-10-20) the author's "
-                      "and translator's notes are restored at the end of the 460 texts that have "
-                      "them: 8,100 author's notes under \"Przypisy autora\" in 337 texts and "
+                      "and translator's notes are restored at the end of the 459 texts that have "
+                      "them: 8,099 author's notes under \"Przypisy autora\" in 336 texts and "
                       "10,838 translator's notes under \"Przypisy tłumacza\" in 168 texts. "
                       "Wolne Lektury's editorial notes are not restored, and the 668 glosses its "
                       "editors added inside 632 of the restored notes (in square brackets, signed "

src/test_attribution_common.py

@@ -26,6 +26,15 @@ from attribution_common import (BUILD_DYNAWORD_STEPS, PERIOD_SPELLING_AT, natura
     (", Henryk", ", Henryk"),
     ("Sienkiewicz,", "Sienkiewicz,"),
     ("", ""),
+    ("Czechowicz,\nJózef", "Józef Czechowicz"),
+    ("Dumas,  Aleksander  (ojciec)", "Dumas, Aleksander (ojciec)"),
+    # the real name behind a pseudonym stays after it, wherever WL puts the bracket
+    ("M-ski, Adam [Zofia Trzeszczkowska]", "Adam M-ski [Zofia Trzeszczkowska]"),
+    ("M-ski, Adam [Zofia  Trzeszczkowska]", "Adam M-ski [Zofia Trzeszczkowska]"),
+    ("M-ski, Adam[Zofia Trzeszczkowska]", "Adam M-ski [Zofia Trzeszczkowska]"),
+    ("M-ski [Zofia Trzeszczkowska], Adam", "Adam M-ski [Zofia Trzeszczkowska]"),
+    ("M-ski, [Zofia Trzeszczkowska]", "M-ski, [Zofia Trzeszczkowska]"),  # only a bracket: unchanged
+    ("Kowalski, Jan [?] Maria", "Jan [?] Maria Kowalski"),  # a bracket inside the given names is not moved
 ])
 def test_natural_order_changes_only_family_comma_given(name, expected):
     assert natural_order(name) == expected

src/test_attribution_release.py

@@ -26,7 +26,7 @@ from sources import SOURCES
 ROOT = Path(__file__).resolve().parents[1]
 Release = namedtuple("Release", "name build table records evidence stats")
 BUILDS = {"wolne_lektury": wl, "eltec_pol": eltec}
-ROWS = {"wolne_lektury": 6076, "eltec_pol": 100}
+ROWS = {"wolne_lektury": 6074, "eltec_pol": 100}
 CREDIT_LINE = "Książka pochodzi z serwisu Wolne Lektury"
 # The full URL of each licence version, written out here and not imported, so that changing the builder's table fails.
 LICENCE_URLS = {
@@ -37,7 +37,7 @@ LICENCE_URLS = {
     "CC-BY-SA-4.0": "https://creativecommons.org/licenses/by-sa/4.0/",
     "CC-BY-4.0": "https://creativecommons.org/licenses/by/4.0/",
 }
-WL_LICENCES = {"public-domain": 4559, "LAL-1.3": 1467, "CC-BY-SA-3.0-PL": 34, "CC-BY-SA-3.0": 5, "CC-BY-SA-4.0": 11}
+WL_LICENCES = {"public-domain": 4557, "LAL-1.3": 1467, "CC-BY-SA-3.0-PL": 34, "CC-BY-SA-3.0": 5, "CC-BY-SA-4.0": 11}
 FORBIDDEN_WORDS = re.compile(r"lawyer|legal advis|solicitor|counsel", re.I)  # we write "our assessment"
 
 
@@ -163,14 +163,15 @@ def test_wl_metadata_is_filled_wherever_wl_has_it(wl_release):
             assert ("Tłumaczenie: " in record["author"]) == bool(record["translators"]), record["id"]
             assert ("Autorzy oryginału: " in record["author"]) == bool(people and record["translators"]), record["id"]
     summary = wl_release.evidence["summary"]
-    assert summary["title_filled"] == summary["rows"] == 6076
+    assert summary["title_filled"] == summary["rows"] == 6074
     assert summary["translator_filled"] == sum(bool(r["translators"]) for r in wl_release.records)
     assert summary["original_title_filled"] == 0
 
 
 def test_the_hand_read_of_the_notes_and_prefaces_sample_refers_to_rows_in_the_sample(wl_release):
-    ids = [r["id"] for r in wl_release.records]
+    ids = sorted([r["id"] for r in wl_release.records] + list(wl.NOT_FREE))  # the rows when the sample was read
     sample = set(wl.notes_sample(ids))
+    assert not sample & wl.NOT_FREE.keys()
     result = wl_release.evidence["notes_and_prefaces"]["hand_sample"]["result"]
     assert len(sample) == wl_release.evidence["notes_and_prefaces"]["hand_sample"]["rows"] == 40
     assert set(result["from_the_print_edition"]) <= sample and result["wl_credit_or_note_text"] == 0
@@ -208,7 +209,7 @@ def test_the_edition_review_of_2026_10_20_covers_98_rows_in_three_disjoint_group
     unnamed = reviewed["no_pd_date_translator_not_named"]
     assert sorted(by_id[i]["title"] for i in unnamed) == ["Cicha noc", "Opowieść wigilijna"] and set(unnamed) <= set(no_date)
     after_1955 = evidence["edition_review"]["ids"]["print_after_1955"]
-    assert reviewed["not_checked"]["rows"] == len(after_1955) == evidence["edition_review"]["flags"]["print_after_1955"] == 1311
+    assert reviewed["not_checked"]["rows"] == len(after_1955) == evidence["edition_review"]["flags"]["print_after_1955"] == 1310
     assert list(reviewed["not_checked"]["wl_date_pd_wrong"]) == ["wolne_lektury_2374660"] and "wolne_lektury_2374660" in critical
 
 
@@ -219,6 +220,23 @@ def test_wl_licensed_records_list_the_people_and_notes_wl_credits(wl_release):
         assert record["credit_notes"] and any(record["credits"].values()), record["id"]
 
 
+def test_the_bn_check_of_2026_10_20_is_in_the_evidence_and_in_the_sidecar(wl_release):
+    evidence, by_id = wl_release.evidence["bn_check"], {r["id"]: r for r in wl_release.records}
+    chosen = wl.restoration(wl.read_notes(), wl_release.table)[0]
+    assert evidence == json.loads(json.dumps(wl.bn_check(wl_release.records, chosen)))
+    assert not wl.NOT_FREE.keys() & by_id.keys() and len(evidence["dropped_as_not_free"]) == 2
+    noted = {i for i, r in by_id.items() if "metadata_note" in r}
+    assert noted == wl.TRANSLATOR_FIXES.keys() | wl.SOURCE_EDITION_NOTES.keys() and len(noted) == 4
+    for row_id, (_, fixed, _) in wl.TRANSLATOR_FIXES.items():
+        assert by_id[row_id]["translators"] == fixed and fixed[-1] in by_id[row_id]["author"], row_id
+    montaigne = evidence["joint_credit_kept"]["rows"]
+    assert [[montaigne[i][k] for k in ("translator_notes", "crediting_cieglewicz", "of_them_greek", "greek_characters",
+                                       "naming_sinko")] for i in sorted(montaigne)] == [
+        [862, 0, 0, 0, 0], [466, 336, 4, 674, 0], [622, 0, 0, 0, 0]]
+    assert evidence["pseudonym_with_real_name"]["rows"] == 22
+    assert not any("Adam [" in t for r in wl_release.records for t in r["translators"])
+
+
 # eltec_pol
 
 def test_every_eltec_record_names_the_eltec_record_the_licence_url_and_the_collection(eltec_release):

src/test_wolne_lektury_attribution.py

@@ -169,6 +169,33 @@ def test_make_record_fails_closed():
         make_record("x", "t", licence(slug="other"), good)
 
 
+def test_make_record_corrects_the_translators_in_translator_fixes_and_says_so(monkeypatch):
+    monkeypatch.setattr(build, "TRANSLATOR_FIXES", {"x": (["Leon Urlich"], ["Leon Ulrich"], "BN record 1."),
+                                                    "y": ([], ["Aniela Zagórska"], "BN record 2.")})
+    fixed = make_record("x", "t", licence(), parse_header(rdf(translators=("Urlich, Leon",))))
+    assert fixed["translators"] == ["Leon Ulrich"]
+    assert fixed["author"] == "Autorzy oryginału: Bolesław Prus; Tłumaczenie: Leon Ulrich"
+    assert fixed["metadata_note"] == ("`translators` corrected from Wolne Lektury's header, which gives \"Leon Urlich\". "
+                                      "BN record 1.")
+    added = make_record("y", "t", licence(), parse_header(rdf()))
+    assert added["translators"] == ["Aniela Zagórska"]
+    assert added["metadata_note"].startswith("`translators` corrected from Wolne Lektury's header, which gives none.")
+    other = make_record("z", "t", licence(), parse_header(rdf(translators=("Urlich, Leon",))))
+    assert other["translators"] == ["Leon Urlich"] and "metadata_note" not in other
+    # WL's header no longer gives the translators the fix was written for
+    with pytest.raises(SystemExit, match="review TRANSLATOR_FIXES"):
+        make_record("x", "t", licence(), parse_header(rdf(translators=("Ulrich, Leon",))))
+
+
+def test_make_record_notes_a_wrong_print_edition_and_joins_it_to_a_translator_fix(monkeypatch):
+    monkeypatch.setattr(build, "SOURCE_EDITION_NOTES", {"x": "Another book.", "y": "Another book."})
+    monkeypatch.setattr(build, "TRANSLATOR_FIXES", {"y": ([], ["Anna Nowak"], "BN record 3.")})
+    noted = make_record("x", "t", licence(), parse_header(rdf()))
+    assert noted["metadata_note"] == "Another book." and noted["source_edition"] == "Bolesław Prus, Lalka, Warszawa 1890."
+    both = make_record("y", "t", licence(), parse_header(rdf()))
+    assert both["metadata_note"].endswith("BN record 3. Another book.")
+
+
 def test_build_records_needs_every_row_and_header():
     table = pa.table({"id": ["a", "b"], "text": ["x", "y"]})
     rows = [licence(row_id="a"), licence(row_id="b", slug="lalka-2")]
@@ -360,9 +387,11 @@ def test_wl_commentary_counts_the_rows_under_a_komentarz_heading():
     assert wl_commentary(records, {})["rows"] == 0
 
 
-def test_the_hand_sample_result_is_recorded_in_the_evidence():
+def test_the_hand_sample_result_is_recorded_in_the_evidence(monkeypatch):
     table = pa.table({"id": ["a"], "text": ["Autor\n\nTytuł\n\nTekst."]})
-    stats = {"drop_short": 0, "drop_lang": 0, "drop_ocr": 0, "drop_dup": 0, "drop_eltec_dup": 63, "drop_rights": 2}
+    stats = {"drop_short": 0, "drop_lang": 0, "drop_ocr": 0, "drop_dup": 0, "drop_eltec_dup": 63, "drop_rights": 2,
+             "drop_not_free": 2}
+    monkeypatch.setattr(build, "bn_check", lambda records, chosen: {})  # it names rows of the released table
     hand = build_evidence(table, [record()], stats, *NO_NOTES)["notes_and_prefaces"]["hand_sample"]
     assert hand["rows"] == NOTES_SAMPLE_SIZE == 40 and hand["result"] == build.NOTES_SAMPLE_RESULT
     assert "450" in hand["read"] and build.NOTES_SAMPLE_RESULT["wl_credit_or_note_text"] == 0
@@ -384,13 +413,15 @@ def test_summary_counts_authors_translators_and_licences():
     assert counts["wl_url_names_another_version"] == 1
 
 
-def test_build_evidence_lists_the_steps_with_their_drop_counts():
+def test_build_evidence_lists_the_steps_with_their_drop_counts(monkeypatch):
     table = pa.table({"id": ["a"], "text": ["Autor\n\nTytuł\n\nTekst."]})
-    stats = {"drop_short": 0, "drop_lang": 0, "drop_ocr": 0, "drop_dup": 0, "drop_eltec_dup": 63, "drop_rights": 2}
+    stats = {"drop_short": 0, "drop_lang": 0, "drop_ocr": 0, "drop_dup": 0, "drop_eltec_dup": 63, "drop_rights": 2,
+             "drop_not_free": 2}
+    monkeypatch.setattr(build, "bn_check", lambda records, chosen: {})  # it names rows of the released table
     evidence = build_evidence(table, [record()], stats, *NO_NOTES)
     steps = evidence["normalisation_steps"]
     assert len(steps) == len(NORMALISATION_STEPS) and all("stat" not in step for step in steps)
-    assert sorted(s["rows_dropped"] for s in steps if "rows_dropped" in s) == [0, 0, 0, 0, 2, 63]
+    assert sorted(s["rows_dropped"] for s in steps if "rows_dropped" in s) == [0, 0, 0, 0, 2, 2, 63]
     assert [s["changes_text"] for s in steps if s["changes_text"] != "no"] == [
         "not known", "leading and trailing whitespace only", "yes, for the rows whose book has them"]
     assert "notes" in [s for s in steps if s["changes_text"].startswith("yes")][0]["step"]
@@ -398,9 +429,54 @@ def test_build_evidence_lists_the_steps_with_their_drop_counts():
     assert evidence["footnote_check"] == footnote_check(*NO_NOTES)
     assert evidence["wl_glosses_in_main_text"] == text_glosses_check()
     assert evidence["text_digest"] == build.TEXT_DIGEST and evidence["summary"]["rows"] == 1
+    assert evidence["bn_check"] == {}
     json.dumps(evidence)  # serialisable
 
 
+def bn_rows(monkeypatch, **named):
+    """Records a, b and c (b translated by Adam M-ski) and BN_CHECK's dicts emptied, except `named`."""
+    for name in ("TRANSLATOR_FIXES", "SOURCE_EDITION_NOTES", "MONTAIGNE", "WL_DATE_PD_EARLY", "WL_RIGHTS_DEATH_YEARS"):
+        monkeypatch.setattr(build, name, named.get(name, {}))
+    monkeypatch.setattr(build, "NOT_FREE", named.get("NOT_FREE", {"gone": "why"}))
+    records = [record(row_id="a"), record(translators=("M-ski, Adam [Zofia Trzeszczkowska]",), row_id="b"),
+               record(row_id="c")]
+    records[0]["wl_rights"] = "Domena publiczna - Jerzy Żuławski zm. 1914"
+    return records
+
+
+def test_bn_check_counts_the_montaigne_credits_and_the_pseudonym(monkeypatch):
+    records = bn_rows(monkeypatch, MONTAIGNE={"a": "Księga pierwsza", "c": "Księga trzecia"},
+                      WL_RIGHTS_DEATH_YEARS={"Jerzy Żuławski zm. 1914": 1915})
+    credit = build.MONTAIGNE_CREDIT
+    chosen = {"a": {"pa": [f"x {credit}"], "pt": [f"Horacy, tłum. {credit}", f"(gr.) --- y; {credit}", "Sinko",
+                                                  "Cicero"]}}
+    check = build.bn_check(records, chosen)
+    assert check["dropped_as_not_free"] == {"gone": "why"}
+    assert check["joint_credit_kept"]["rows"] == {
+        "a": {"book": "Księga pierwsza", "translator_notes": 4, "crediting_cieglewicz": 2, "of_them_greek": 1,
+              "greek_characters": len(f"(gr.) --- y; {credit}"), "naming_sinko": 1},
+        "c": {"book": "Księga trzecia", "translator_notes": 0, "crediting_cieglewicz": 0, "of_them_greek": 0,
+              "greek_characters": 0, "naming_sinko": 0}}
+    assert check["pseudonym_with_real_name"]["rows"] == 1
+    assert check["kept_as_wolne_lektury_wrote"]["death_year_in_wl_rights"] == {
+        "Jerzy Żuławski zm. 1914": {"bn_death_year": 1915, "ids": ["a"]}}
+    json.dumps(check)
+
+
+@pytest.mark.parametrize("named", [
+    {"TRANSLATOR_FIXES": {"gone": ([], [], "")}}, {"SOURCE_EDITION_NOTES": {"gone": ""}},
+    {"MONTAIGNE": {"gone": ""}}, {"WL_DATE_PD_EARLY": {"gone": ""}}, {"NOT_FREE": {"a": "why"}}])
+def test_bn_check_fails_on_a_row_not_in_the_table_or_a_kept_not_free_row(monkeypatch, named):
+    with pytest.raises(SystemExit, match="bn_check names rows"):
+        build.bn_check(bn_rows(monkeypatch, **named), {})
+
+
+def test_bn_check_fails_on_a_death_year_text_in_no_rights_line(monkeypatch):
+    records = bn_rows(monkeypatch, WL_RIGHTS_DEATH_YEARS={"Jerzy Żuławski zm. 1914": 1915, "Nikt zm. 1900": 1901})
+    with pytest.raises(SystemExit, match="WL_RIGHTS_DEATH_YEARS"):
+        build.bn_check(records, {})
+
+
 def test_footnote_check_counts_what_was_restored_and_lists_what_was_left_out():
     chosen = {"a": {"pa": ["Jeden.", "Dwa."]}, "b": {"pa": ["Trzy."], "pt": ["Cztery."]}}
     left_out = {"empty": [("c", "pa", 0)], "in_text": [], "translator_without_pd_date": [("d", "pt", 3)],

src/test_wolne_lektury_contract.py

@@ -16,11 +16,10 @@ sys.path.insert(0, str(Path(__file__).resolve().parent))
 
 from attribution_common import text_digest
 from build_eltec_pol_attribution import TEXT_DIGEST as ELTEC_TEXT_DIGEST
-from clean_wolne_lektury import (DATA, DROP_AT, ELTEC, PAIRS, REPORT_AT, build, drop_unlicensed,
-                                 relabel, wl_title)
+from clean_wolne_lektury import DATA, DROP_AT, ELTEC, PAIRS, REPORT_AT, build, drop_listed, relabel, wl_title
 from shingle_containment import best_containment, strip_wl_footer
 from sources import SOURCES
-from wolne_lektury_licences import LABELS, LICENCES, MANUAL, PUBLIC_DOMAIN, UNLICENSED
+from wolne_lektury_licences import LABELS, LICENCES, MANUAL, NOT_FREE, PUBLIC_DOMAIN, UNLICENSED
 
 # Hand-checked: only part of each is in its ELTeC novel, so they stay (see the findings entry).
 PARTIAL_IDS = {"wolne_lektury_2375475", "wolne_lektury_2375476", "wolne_lektury_2376242"}
@@ -47,15 +46,16 @@ def test_release_has_no_eltec_duplicates_and_matches_stats():
     dropped = {p["wl_id"] for p in pairs if p["action"] == "drop"}
     partial = {p["wl_id"] for p in pairs if p["action"] == "keep_partial"}
 
-    assert table.num_rows == len(ids) == stats["kept"] == 6_076
+    assert table.num_rows == len(ids) == stats["kept"] == 6_074
     assert len(dropped) == stats["drop_eltec_dup"] == 63
     assert len(UNLICENSED) == stats["drop_rights"] == 2 and not UNLICENSED & ids
+    assert len(NOT_FREE) == stats["drop_not_free"] == 2 and not NOT_FREE.keys() & ids
     assert partial == PARTIAL_IDS
     assert not dropped & ids and partial <= ids
     assert all(float(p["score"]) >= DROP_AT for p in pairs if p["action"] == "drop")
     assert all(REPORT_AT <= float(p["score"]) < DROP_AT for p in pairs if p["action"] == "keep_partial")
     assert all(p["eltec_id"].startswith("eltec_pol_") for p in pairs)
-    drops = ("drop_short", "drop_lang", "drop_dup", "drop_ocr", "drop_eltec_dup", "drop_rights")
+    drops = ("drop_short", "drop_lang", "drop_dup", "drop_ocr", "drop_eltec_dup", "drop_rights", "drop_not_free")
     assert stats["read"] - sum(stats[k] for k in drops) == stats["kept"]
     assert pc.sum(table.column("token_count")).as_py() == stats["tokens"]
     assert pc.sum(pc.utf8_length(table.column("text"))).as_py() == stats["chars"]
@@ -111,11 +111,12 @@ def test_every_row_carries_its_own_wolne_lektury_licence():
     assert tome["license"] == PUBLIC_DOMAIN and "Boy-Żeleński" in tome["wl_rights"]
 
 
-def test_drop_unlicensed_fails_closed():
-    rows = pa.table({"id": ["a", *sorted(UNLICENSED)]})
-    assert drop_unlicensed(rows).column("id").to_pylist() == ["a"]
-    with pytest.raises(SystemExit, match="UNLICENSED"):
-        drop_unlicensed(rows.slice(0, 2))  # one UNLICENSED id is missing
+@pytest.mark.parametrize("ids, name", [(UNLICENSED, "UNLICENSED"), (NOT_FREE, "NOT_FREE")])
+def test_drop_listed_fails_closed(ids, name):
+    rows = pa.table({"id": ["a", *sorted(ids)]})
+    assert drop_listed(rows, ids, name).column("id").to_pylist() == ["a"]
+    with pytest.raises(SystemExit, match=name):
+        drop_listed(rows.slice(0, 2), ids, name)  # one listed id is missing
 
 
 def test_relabel_fails_closed(tmp_path):
@@ -138,6 +139,7 @@ def test_datasheet_counts_match_stats():
     assert f"| {stats['kept']:,} | {stats['chars']:,} | {stats['tokens']:,} |" in card
     assert f"{stats['drop_eltec_dup']} " in card and "eltec_pol" in card
     assert f"{stats['drop_rights']} documents" in card and "Kozioł" in card
+    assert f"{stats['drop_not_free']} more are dropped" in card and "Frycz (d. 1960)" in card and "born in 1932" in card
     assert f"; {len(MANUAL)} were matched by hand" in card
     for label, count in stats["licenses"].items():
         assert f"| `{label}` | {count:,} |" in card

src/test_wolne_lektury_licences.py

@@ -170,6 +170,16 @@ def test_unlicensed_rows_are_left_out_until_they_label_cleanly(tmp_path, monkeyp
     assert problems == ["now-clean: in UNLICENSED but labels as {'public-domain'}"]
 
 
+def test_not_free_rows_are_left_out_while_they_label_public_domain(tmp_path, monkeypatch):
+    cache = _cache(tmp_path, {"to-lubie": (PART, "Domena publiczna - Adam Mickiewicz zm. 1855"),
+                              "ballady": (COLLECTION, "Licencja Wolnej Sztuki 1.3")})
+    monkeypatch.setattr(wolne_lektury_licences, "NOT_FREE", {"joint": "why", "relicensed": "why"})
+    rows = [{"id": "joint", "text": PART}, {"id": "relicensed", "text": COLLECTION}]
+    out, problems = label_rows(rows, cache)
+    assert out == []
+    assert problems == ["relicensed: in NOT_FREE but labels as {'LAL-1.3'}"]
+
+
 def test_fetch_takes_media_urls_from_the_api_entry(tmp_path, monkeypatch):
     for part in ("hdr", "head", "tail"):
         (tmp_path / part).mkdir()

src/test_wolne_lektury_notes.py

@@ -481,11 +481,11 @@ def test_the_notes_file_is_the_pinned_one_and_lf_json_lines_sorted_by_licences_o
     assert all(re.fullmatch(r"[0-9a-f]{64}", r["xml_sha256"]) and (r["pa"] or r["pt"]) for r in rows)
 
 
-def test_the_notes_file_has_460_rows_and_the_notes_counted_in_the_evidence(table_and_notes):
+def test_the_notes_file_has_459_rows_and_the_notes_counted_in_the_evidence(table_and_notes):
     _, notes = table_and_notes
-    assert len(notes) == 460
-    assert sum(len(r["pa"]) for r in notes.values()) == 8105 and sum(len(r["pt"]) for r in notes.values()) == 10838
-    assert sum(bool(r["pa"]) for r in notes.values()) == 337 + 0 and sum(bool(r["pt"]) for r in notes.values()) == 168
+    assert len(notes) == 459
+    assert sum(len(r["pa"]) for r in notes.values()) == 8104 and sum(len(r["pt"]) for r in notes.values()) == 10838
+    assert sum(bool(r["pa"]) for r in notes.values()) == 336 + 0 and sum(bool(r["pt"]) for r in notes.values()) == 168
 
 
 def test_the_notes_file_holds_the_licences_slug_of_each_row_and_wl_s_public_domain_date():
@@ -503,7 +503,7 @@ def test_the_released_text_minus_the_appended_sections_is_the_text_before_the_no
     texts = released_table.column("text").to_pylist()
     before = [t[:appended_start(t)] if i in chosen else t for i, t in zip(ids, texts, strict=True)]
     assert text_digest(released_table.set_column(1, "text", pa.array(before, pa.string()))) == TEXT_DIGEST_WITHOUT_GLOSSES
-    # the digest covers the 5,616 rows without notes too: they are byte-identical to the text the cleaning had before
+    # the digest covers the 5,615 rows without notes too: they are byte-identical to the text the cleaning had before
     # the notes (TEXT_DIGEST_BEFORE, minus the two glosses of TEXT_GLOSSES; src/clean_wolne_lektury.py checks both)
 
 
@@ -522,7 +522,7 @@ def test_each_changed_row_ends_with_exactly_its_sections_and_no_row_has_a_footer
     chosen, _, _ = restoration(notes, released_table)
     ids = released_table.column("id").to_pylist()
     texts = dict(zip(ids, released_table.column("text").to_pylist(), strict=True))
-    assert len(chosen) == 460 and set(chosen) <= set(notes)
+    assert len(chosen) == 459 and set(chosen) <= set(notes)
     for row_id, kinds in chosen.items():
         text = texts[row_id]
         assert text.endswith("\n\n" + sections(kinds)), row_id
@@ -556,7 +556,7 @@ def test_each_section_is_numbered_from_one_and_holds_the_notes_of_the_notes_file
             parsed[tag].append(note)
         assert parsed == expected, row_id
         changed += 1
-    assert changed == 460
+    assert changed == 459
 
 
 def test_a_section_heading_is_not_added_for_a_kind_a_row_has_no_notes_of(table_and_notes):
@@ -586,9 +586,9 @@ def test_the_token_counts_of_the_changed_rows_are_recounted_and_the_stats_add_up
     tokens = released_table.column("token_count").to_pylist()
     ids = released_table.column("id").to_pylist()
     changed = [k for k, i in enumerate(ids) if i in notes or i in TEXT_GLOSSES]
-    assert len(changed) == stats["notes_rows"] + len(TEXT_GLOSSES) == 460 + 2
-    # ponytail: the other 5,616 token counts are those of v0.2.5, whose texts the digest test pins; recounting
-    # all 6,076 texts takes minutes
+    assert len(changed) == stats["notes_rows"] + len(TEXT_GLOSSES) == 459 + 2
+    # ponytail: the other 5,613 token counts are those of v0.2.5, whose texts the digest test pins; recounting
+    # all 6,074 texts takes minutes
     assert wn.token_counts([texts[k] for k in changed]) == [tokens[k] for k in changed]
     assert sum(map(len, texts)) == stats["chars"] and sum(tokens) == stats["tokens"]
     assert stats["notes_sha256"] == NOTES_SHA256 and stats["kept"] == len(ids)
@@ -602,7 +602,7 @@ def test_the_evidence_footnote_check_is_what_the_notes_file_and_the_table_give(t
     assert evidence["footnote_check"] == json.loads(json.dumps(build.footnote_check(chosen, left_out, stripped)))
     check = evidence["footnote_check"]
     assert check["status"] == "restored" and check["notes_file_sha256"] == NOTES_SHA256 == sha256_file(NOTES)
-    assert check["restored"] == {"pa": {"rows": 337, "notes": 8100, "characters": 1854770},
+    assert check["restored"] == {"pa": {"rows": 336, "notes": 8099, "characters": 1854631},
                                  "pt": {"rows": 168, "notes": 10838, "characters": 1923101}}
     assert [len(check["left_out"][k]["notes"]) for k in ("empty", "in_text", "translator_without_pd_date",
                                                          "wl_gloss_only")] == [2, 3, 0, 0]

src/wolne_lektury_licences.py

@@ -86,6 +86,21 @@ UNLICENSED = {
     # Horace, tr. Paweł Kozioł: "CC BY-SA" with no version or URL; its WL footer says public domain
     "wolne_lektury_2373208",  # do-taliarcha
 }
+# Rows that Wolne Lektury marks public domain although their translation is not free, found by the check against the
+# National Library's catalogue (BN) and dropped by the maintainer's decision of 2026-10-20 (polish-dynaword-v2#109).
+# They are left out like UNLICENSED, and src/clean_wolne_lektury.py drops them. A row here that no longer labels public
+# domain fails the run, so the drop is reviewed again.
+NOT_FREE = {
+    "wolne_lektury_2375754": "Nietzsche, \"Wola mocy\" (nietzsche-wola-mocy): a joint translation by Stefan Frycz and "
+                             "Konrad Drzewiecki (BN record 1826277, Warszawa: J. Mortkowicz, 1911: \"przeł. Stefan Frycz "
+                             "i Konrad Drzewiecki\"; \"Frycz, Stefan (1876-1960)\"). Wolne Lektury counts Drzewiecki "
+                             "alone (d. 1922); a joint work is protected for 70 years after the last co-author's death, "
+                             "so to the end of 2030.",
+    "wolne_lektury_2371705": "Wells, \"Wehikuł czasu\" (wells-wehikul-czasu): transcribed from the Ossolineum print "
+                             "(Wrocław 1985), whose translation was revised by an editor born in 1932 (BN record 1216473: "
+                             "\"przeł. [z ang.] Feliks Wermiński ; przekł. zweryfikował i oprac. Juliusz K. Palczewski\"; "
+                             "\"Palczewski, Juliusz K. (1932- )\"). Wolne Lektury credits Wermiński (d. 1917) only.",
+}
 
 
 def normalise(text: str) -> str:
@@ -248,6 +263,10 @@ def label_rows(rows: list[dict], cache: Path) -> tuple[list[dict], list[str]]:
             found[slug] = (classify(rights, urls), rights, " ".join(urls))
         labels = {label for label, _, _ in found.values()}
         unusable = not hits or None in labels or len(labels) > 1  # identical texts must agree on the licence
+        if row["id"] in NOT_FREE:
+            if labels != {PUBLIC_DOMAIN}:
+                problems.append(f"{row['id']}: in NOT_FREE but labels as {labels}")
+            continue
         if row["id"] in UNLICENSED:
             if not unusable:
                 problems.append(f"{row['id']}: in UNLICENSED but labels as {labels}")

src/wolne_lektury_notes.py

@@ -63,7 +63,7 @@ from shingle_containment import strip_wl_footer  # noqa: E402
 from wolne_lektury_licences import DATA, PUBLIC_DOMAIN, read_licences  # noqa: E402
 
 NOTES = DATA / "wolne_lektury.notes.jsonl"
-NOTES_SHA256 = "07fe95a8265f6b40512305cf39b4960fbd16ec2ba9c4c68f7ce92b681d5a51a1"
+NOTES_SHA256 = "da18d6a7eebf88f63efbe1e082d5e2cdc387bf9bbf808cc9d0a6411dbf14d8e1"
 HEADINGS = {"pa": "Przypisy autora", "pt": "Przypisy tłumacza"}  # in this order
 OWNER = {"pa": "author's", "pt": "translator's"}
 LEFT_OUT = {"pe", "pr", "extra", "uwaga"}  # WL's editorial notes and remarks, never part of a note
@@ -71,14 +71,15 @@ DC_DATE_PD = "{http://purl.org/dc/elements/1.1/}date.pd"
 PARAGRAPH = "\n\n"
 NO_XML = {"korczak-jak-kochac-dziecko-dziecko-w-rodzinie"}  # WL serves no XML for this book (404, 2026-10-20)
 # text_digest of the wolne_lektury table before the glosses and the notes: the clean output of the v0.2.5 Parquet at
-# 6b7886e
-TEXT_DIGEST_BEFORE = "f4973619094cd415a9a49f44b381ed6c45d5c390f4c58d999819957cf2f25f88"
+# 6b7886e (f4973619094cd415a9a49f44b381ed6c45d5c390f4c58d999819957cf2f25f88) without the two NOT_FREE rows of
+# src/wolne_lektury_licences.py
+TEXT_DIGEST_BEFORE = "578bd139e7c4a064713aeaf5a7d7a283a08814f57cb25214e5c0f775d715d433"
 # The glosses of WL's editors in the main text: what without_glosses() finds in the table above (2026-10-20).
 # strip_text_glosses() fails unless it finds exactly these.
 TEXT_GLOSSES = {"wolne_lektury_2372131": ["[słońce – przyp. WL]"],
                 "wolne_lektury_2377006": ["[mówi jednocześnie z dalszymi kwestiami Millera i Ferdynanda — Red.WL]"]}
 # text_digest of that table after strip_text_glosses(): the text the notes are added to
-TEXT_DIGEST_WITHOUT_GLOSSES = "09f4ab3eb640b78d83ce26230a7efa8bc10ff88a396f588dd38ba7c1cbb3d54f"
+TEXT_DIGEST_WITHOUT_GLOSSES = "45fe3369c6b7f509e2c84cfff161ce251da3963ee9f972cd38da861f7e409eb7"
 # (row id, tag, position among the book's notes of that tag) of the notes the export kept in the text: the whole note
 # is in the row's text and not in the book's text outside its notes (2026-10-20). A note that the main text only
 # quotes, or paraphrases in brackets (the four Lithuanian glosses of wolne_lektury_2376761), is not the note: restored.