tamnd commited on
Commit
4d58f4c
·
verified ·
1 Parent(s): 30a21a6

Publish shard CC-MAIN-2026-08/00282

Browse files
Files changed (3) hide show
  1. README.md +11 -11
  2. data/CC-MAIN-2026-08/00282.parquet +3 -0
  3. stats.csv +2 -1
README.md CHANGED
@@ -32,7 +32,7 @@ configs:
32
 
33
  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.
34
 
35
- The dataset currently includes crawl **CC-MAIN-2026-08** with **3,593,863 documents across 184 shards**. We plan to add more snapshots over time.
36
 
37
  Open Index is released under the **Open Data Commons Attribution License (ODC-By) v1.0**, the same license used by Common Crawl.
38
 
@@ -166,27 +166,27 @@ Empty conversions (pages where trafilatura could not extract meaningful content)
166
 
167
  ### Compression Ratios
168
 
169
- Numbers below are actual measurements summed across all 184 files of CC-MAIN-2026-08 (3,593,863 pages total), projected to the full crawl of 100,000 WARC files.
170
 
171
- | Stage | 184 files (measured) | 100,000 files (projected) | Reduction |
172
  |---|---|---|---|
173
- | Raw WARC (.warc.gz, downloaded) | ~149.1 GB | ~83 TB | — |
174
- | HTML extracted (uncompressed) | 445.0 GB | ~295 TB | — |
175
- | Packed markdown WARC (.md.warc.gz) | ~7.5 GB | ~3.7 TB | **-98.3%** vs HTML |
176
  | Final Parquet (Zstd level 19) | 5.1 GB | ~2.9 TB | **-32.1%** vs packed WARC |
177
 
178
- The big win is the HTML → Markdown step: trafilatura strips all tags, scripts, styles, navigation, and ads, keeping only the main content. This cuts 445.0 GB of uncompressed HTML down to 16.0 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%.
179
 
180
- End to end: ~149.1 GB of raw gzipped WARCs becomes **5.1 GB of Parquet** — a **96.6% total reduction** — containing 3,593,863 clean markdown documents.
181
 
182
  ### Processing Times
183
 
184
- Pipeline timings across 184 shards of CC-MAIN-2026-08:
185
 
186
  ```
187
- Download (raw WARC) ████████████████████████ total 6h 4m 1s avg 1m 58s
188
  Convert (HTML → MD) █░░░░░░░░░░░░░░░░░░░░░░░ total 16m 2s avg 5s
189
- Export (Parquet) ████░░░░░░░░░░░░░░░░░░░░ total 1h 14m 12s avg 24s
190
  Publish (HuggingFace) ███████████░░░░░░░░░░░░░ total 2h 57m 20s avg 57s
191
  ```
192
 
 
32
 
33
  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.
34
 
35
+ The dataset currently includes crawl **CC-MAIN-2026-08** with **3,613,288 documents across 185 shards**. We plan to add more snapshots over time.
36
 
37
  Open Index is released under the **Open Data Commons Attribution License (ODC-By) v1.0**, the same license used by Common Crawl.
38
 
 
166
 
167
  ### Compression Ratios
168
 
169
+ Numbers below are actual measurements summed across all 185 files of CC-MAIN-2026-08 (3,613,288 pages total), projected to the full crawl of 100,000 WARC files.
170
 
171
+ | Stage | 185 files (measured) | 100,000 files (projected) | Reduction |
172
  |---|---|---|---|
173
+ | Raw WARC (.warc.gz, downloaded) | ~150.0 GB | ~83 TB | — |
174
+ | HTML extracted (uncompressed) | 447.4 GB | ~295 TB | — |
175
+ | Packed markdown WARC (.md.warc.gz) | ~7.6 GB | ~3.7 TB | **-98.3%** vs HTML |
176
  | Final Parquet (Zstd level 19) | 5.1 GB | ~2.9 TB | **-32.1%** vs packed WARC |
177
 
178
+ The big win is the HTML → Markdown step: trafilatura strips all tags, scripts, styles, navigation, and ads, keeping only the main content. This cuts 447.4 GB of uncompressed HTML down to 16.1 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%.
179
 
180
+ End to end: ~150.0 GB of raw gzipped WARCs becomes **5.1 GB of Parquet** — a **96.6% total reduction** — containing 3,613,288 clean markdown documents.
181
 
182
  ### Processing Times
183
 
184
+ Pipeline timings across 185 shards of CC-MAIN-2026-08:
185
 
186
  ```
187
+ Download (raw WARC) ████████████████████████ total 6h 5m 28s avg 1m 58s
188
  Convert (HTML → MD) █░░░░░░░░░░░░░░░░░░░░░░░ total 16m 2s avg 5s
189
+ Export (Parquet) ████░░░░░░░░░░░░░░░░░░░░ total 1h 14m 33s avg 24s
190
  Publish (HuggingFace) ███████████░░░░░░░░░░░░░ total 2h 57m 20s avg 57s
191
  ```
192
 
data/CC-MAIN-2026-08/00282.parquet ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:becd94b58569fab933a81a8fcef8e716f48cc3516ad113f845497b44614944ac
3
+ size 29892625
stats.csv CHANGED
@@ -109,6 +109,7 @@ CC-MAIN-2026-08,278,19322,2550292792,95398249,30935270,2026-03-13T08:16:50Z,98,0
109
  CC-MAIN-2026-08,279,19479,2561365029,94152406,30402050,2026-03-13T08:19:10Z,101,0,23,14
110
  CC-MAIN-2026-08,280,19648,2618355119,95881188,30182638,2026-03-13T08:23:20Z,154,0,24,71
111
  CC-MAIN-2026-08,281,19416,2587645863,94698501,30673039,2026-03-13T08:26:40Z,136,0,23,38
 
112
  CC-MAIN-2026-08,401,19517,2605589961,95609438,30404753,2026-03-13T06:41:08Z,218,0,47,107
113
  CC-MAIN-2026-08,402,19446,2552785517,96654791,30588984,2026-03-13T06:43:58Z,98,0,26,44
114
  CC-MAIN-2026-08,403,19320,2579048144,91406292,29375992,2026-03-13T06:49:29Z,198,0,33,98
@@ -182,4 +183,4 @@ CC-MAIN-2026-08,875,19632,2619435387,96015289,30445181,2026-03-13T08:15:18Z,83,0
182
  CC-MAIN-2026-08,876,19247,2577843957,92441652,29314063,2026-03-13T08:19:04Z,94,0,31,100
183
  CC-MAIN-2026-08,877,19333,2553485150,91901591,29616717,2026-03-13T08:22:56Z,108,0,46,75
184
  CC-MAIN-2026-08,878,19371,2591184192,98118180,30281755,2026-03-13T08:25:59Z,147,0,28,6
185
- CC-MAIN-2026-08,879,19553,2569444239,95701279,30858541,2026-03-13T08:27:52Z,88,0,23,0
 
109
  CC-MAIN-2026-08,279,19479,2561365029,94152406,30402050,2026-03-13T08:19:10Z,101,0,23,14
110
  CC-MAIN-2026-08,280,19648,2618355119,95881188,30182638,2026-03-13T08:23:20Z,154,0,24,71
111
  CC-MAIN-2026-08,281,19416,2587645863,94698501,30673039,2026-03-13T08:26:40Z,136,0,23,38
112
+ CC-MAIN-2026-08,282,19425,2542414332,94704569,29892625,2026-03-13T08:28:30Z,87,0,21,0
113
  CC-MAIN-2026-08,401,19517,2605589961,95609438,30404753,2026-03-13T06:41:08Z,218,0,47,107
114
  CC-MAIN-2026-08,402,19446,2552785517,96654791,30588984,2026-03-13T06:43:58Z,98,0,26,44
115
  CC-MAIN-2026-08,403,19320,2579048144,91406292,29375992,2026-03-13T06:49:29Z,198,0,33,98
 
183
  CC-MAIN-2026-08,876,19247,2577843957,92441652,29314063,2026-03-13T08:19:04Z,94,0,31,100
184
  CC-MAIN-2026-08,877,19333,2553485150,91901591,29616717,2026-03-13T08:22:56Z,108,0,46,75
185
  CC-MAIN-2026-08,878,19371,2591184192,98118180,30281755,2026-03-13T08:25:59Z,147,0,28,6
186
+ CC-MAIN-2026-08,879,19553,2569444239,95701279,30858541,2026-03-13T08:28:36Z,88,0,23,43