Publish shard CC-MAIN-2026-08/00278

README.md

@@ -32,7 +32,7 @@ configs:
 
 Open Index is a large-scale web text dataset built from [Common Crawl](https://commoncrawl.org). Every page goes through a pipeline that extracts the main content from raw HTML, converts it to clean Markdown using [trafilatura](https://github.com/adbar/trafilatura), and packages the result into Parquet files with full WARC metadata preserved.
 
-The dataset currently includes crawl **CC-MAIN-2026-08** with **3,283,854 documents across 168 shards**. We plan to add more snapshots over time.
+The dataset currently includes crawl **CC-MAIN-2026-08** with **3,322,578 documents across 170 shards**. We plan to add more snapshots over time.
 
 Open Index is released under the **Open Data Commons Attribution License (ODC-By) v1.0**, the same license used by Common Crawl.
 
@@ -166,28 +166,28 @@ Empty conversions (pages where trafilatura could not extract meaningful content)
 
 ### Compression Ratios
 
-Numbers below are actual measurements summed across all 168 files of CC-MAIN-2026-08 (3,283,854 pages total), projected to the full crawl of 100,000 WARC files.
+Numbers below are actual measurements summed across all 170 files of CC-MAIN-2026-08 (3,322,578 pages total), projected to the full crawl of 100,000 WARC files.
 
-| Stage | 168 files (measured) | 100,000 files (projected) | Reduction |
+| Stage | 170 files (measured) | 100,000 files (projected) | Reduction |
 |---|---|---|---|
-| Raw WARC (.warc.gz, downloaded) | ~136.2 GB | ~83 TB | — |
-| HTML extracted (uncompressed) | 406.8 GB | ~295 TB | — |
+| Raw WARC (.warc.gz, downloaded) | ~137.8 GB | ~83 TB | — |
+| HTML extracted (uncompressed) | 411.5 GB | ~295 TB | — |
 | Packed markdown WARC (.md.warc.gz) | ~6.9 GB | ~3.7 TB | **-98.3%** vs HTML |
 | Final Parquet (Zstd level 19) | 4.7 GB | ~2.9 TB | **-32.0%** vs packed WARC |
 
-The big win is the HTML → Markdown step: trafilatura strips all tags, scripts, styles, navigation, and ads, keeping only the main content. This cuts 406.8 GB of uncompressed HTML down to 14.6 GB of markdown — a **98.3% reduction** — before any file-level compression is applied. Parquet with Zstd level 19 then compresses the markdown a further 68.1%.
+The big win is the HTML → Markdown step: trafilatura strips all tags, scripts, styles, navigation, and ads, keeping only the main content. This cuts 411.5 GB of uncompressed HTML down to 14.8 GB of markdown — a **98.3% reduction** — before any file-level compression is applied. Parquet with Zstd level 19 then compresses the markdown a further 68.1%.
 
-End to end: ~136.2 GB of raw gzipped WARCs becomes **4.7 GB of Parquet** — a **96.6% total reduction** — containing 3,283,854 clean markdown documents.
+End to end: ~137.8 GB of raw gzipped WARCs becomes **4.7 GB of Parquet** — a **96.6% total reduction** — containing 3,322,578 clean markdown documents.
 
 ### Processing Times
 
-Pipeline timings across 168 shards of CC-MAIN-2026-08:
+Pipeline timings across 170 shards of CC-MAIN-2026-08:
 
 ```
-Download (raw WARC)        ████████████████████████  total 5h 33m 8s     avg 1m 58s
+Download (raw WARC)        ████████████████████████  total 5h 36m 19s    avg 1m 58s
 Convert  (HTML → MD)       █░░░░░░░░░░░░░░░░░░░░░░░  total 16m 2s        avg 5s
-Export   (Parquet)          ████░░░░░░░░░░░░░░░░░░░░  total 1h 7m 40s     avg 24s
-Publish  (HuggingFace)      █████████░░░░░░░░░░░░░░░  total 2h 8m 10s     avg 45s
+Export   (Parquet)          ████░░░░░░░░░░░░░░░░░░░░  total 1h 8m 30s     avg 24s
+Publish  (HuggingFace)      ███████████░░░░░░░░░░░░░  total 2h 45m 45s    avg 58s
 ```
 
 ### Dataset Charts

data/CC-MAIN-2026-08/00278.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54a2123f9df2d5397dfa766e3cf087e512998875584477aca1b64f451d7a5e04
+size 30935270

stats.csv

@@ -73,6 +73,7 @@ CC-MAIN-2026-08,123,19514,2594071373,96422892,29984857,2026-03-13T08:04:05Z,100,
 CC-MAIN-2026-08,124,19516,2599820114,96915408,30891980,2026-03-13T08:09:10Z,100,0,28,175
 CC-MAIN-2026-08,125,19254,2558572277,96621435,30532558,2026-03-13T08:12:05Z,80,0,20,73
 CC-MAIN-2026-08,126,19346,2586860649,94975987,31506543,2026-03-13T08:14:01Z,84,0,22,9
+CC-MAIN-2026-08,127,19402,2552000966,96535361,31142501,2026-03-13T08:16:03Z,93,0,27,0
 CC-MAIN-2026-08,251,19610,2594898345,97558492,31917087,2026-03-13T06:42:12Z,277,0,23,136
 CC-MAIN-2026-08,252,19240,2525928701,98163046,31104034,2026-03-13T06:47:10Z,238,0,49,10
 CC-MAIN-2026-08,253,19312,2557230855,96017199,30399966,2026-03-13T06:51:42Z,130,0,23,117
@@ -100,6 +101,7 @@ CC-MAIN-2026-08,274,19329,2536901017,91117521,28952785,2026-03-13T08:07:56Z,96,0
 CC-MAIN-2026-08,275,19574,2625560972,98453515,30843942,2026-03-13T08:10:57Z,75,0,20,83
 CC-MAIN-2026-08,276,19412,2592300808,92778825,29186554,2026-03-13T08:12:45Z,81,0,19,6
 CC-MAIN-2026-08,277,19420,2584694709,96412947,30707468,2026-03-13T08:14:39Z,84,0,22,6
+CC-MAIN-2026-08,278,19322,2550292792,95398249,30935270,2026-03-13T08:16:42Z,98,0,23,0
 CC-MAIN-2026-08,401,19517,2605589961,95609438,30404753,2026-03-13T06:41:08Z,218,0,47,107
 CC-MAIN-2026-08,402,19446,2552785517,96654791,30588984,2026-03-13T06:43:58Z,98,0,26,44
 CC-MAIN-2026-08,403,19320,2579048144,91406292,29375992,2026-03-13T06:49:29Z,198,0,33,98
@@ -115,7 +117,7 @@ CC-MAIN-2026-08,412,19421,2588662566,94735501,31063646,2026-03-13T07:27:01Z,121,
 CC-MAIN-2026-08,413,19245,2557370915,94249339,29958343,2026-03-13T07:30:13Z,144,0,31,15
 CC-MAIN-2026-08,414,19623,2591080218,95450599,29779422,2026-03-13T07:33:40Z,103,0,22,81
 CC-MAIN-2026-08,415,19293,2533466692,97370528,31311143,2026-03-13T07:36:14Z,109,0,31,11
-CC-MAIN-2026-08,416,19548,2610354746,98356232,30913041,2026-03-13T07:38:23Z,108,0,20,0
+CC-MAIN-2026-08,416,19548,2610354746,98356232,30913041,2026-03-13T08:15:49Z,108,0,20,2246
 CC-MAIN-2026-08,551,19551,2609034951,89980913,29167227,2026-03-13T06:42:51Z,280,0,20,176
 CC-MAIN-2026-08,552,19402,2594817630,94092761,30309254,2026-03-13T06:48:30Z,211,0,46,80
 CC-MAIN-2026-08,553,19856,2617133708,99106486,30365588,2026-03-13T06:52:51Z,116,0,23,120