The crawl archive for May 2017 is now available! The archive is located in the commoncrawl bucket at crawl-data/CC-MAIN-2017-22/. It contains 2.96 billion+ web pages and over 250 TiB of uncompressed content.
To improve coverage and freshness we used the top 25 million ranked hosts from the Feb/Mar/Apr 2017 webgraph data set and added about 500 million new URLs (not contained in any crawl archive before), of which:
- 330 million URLs were found by a side crawl within a maximum of 3 links (“hops”) away from the home pages of the top 25 million hosts;
- 160 million URLs are a random sample extracted from sitemaps (if provided by any of these 25 million hosts).
About 40% of the crawl archive's 2.96 billion URLs overlap with the preceding April 2017 crawl. The following changes have been made to WARC (also WAT and WET) files:
- the timestamp in WARC filenames now indicates the capture time (fetch time) of the WARC content (see details)
- WARC files and the URL index now contain the detected MIME type (based on the actual content) in addition to the "Content-Type" sent in the HTTP response (details).
To assist with exploring and using the dataset, we provide gzipped files which list all segments, WARC, WAT and WET files. By simply adding either s3://commoncrawl/ or https://data.commoncrawl.org/ to each line, you end up with the S3 and HTTP paths respectively.
The Common Crawl URL Index for this crawl is available at: https://index.commoncrawl.org/CC-MAIN-2017-22/. For more information on working with the URL index, please refer to the previous blog post or the Index Server API. There is also a command-line tool client for common use cases of the URL index.
Please donate to Common Crawl if you appreciate our free datasets! We’re also seeking corporate sponsors to partner with Common Crawl for our non-profit work in open data. Please contact info@commoncrawl.org for sponsorship information.
Erratum:
Unreliable truncation indicator in WARC records
Due to an issue with our crawler, not all truncations were flagged correctly in the WARC headers. A workaround to detect length truncation is to be suspicious if the length of the content is exactly 1048576 bytes. Truncations for time or network do not have such a workaround.
Erratum:
Missing content_truncated flag in URL indexes
The flag in our URL indexes (CDX and columnar) that indicates whether or not a WARC record payload was truncated was added in CC-MAIN-2019-47. This indicator is missing in our indexes for all previous crawl releases.
For more information please refer to the blog post announcing the November 2019 crawl. The reason for the truncation is given only for truncated records following the WARC header field "WARC-Truncated".
Erratum:
SURT URLs do not properly encode non-UTF-8 percent-encoded characters
When constructing SURT (Sort-friendly URI Reordering Transform) URLs, percent-encoded characters that are not valid UTF-8 sequences were not being correctly handled. This could lead to inconsistencies in URL normalization and sorting, potentially causing incorrect deduplication or retrieval issues in datasets that rely on SURT-based indexing. This was addressed in commoncrawl/nutch@6b2d9ea.
Erratum:
WAT data: repeated WARC and HTTP headers are not preserved
Repeated HTTP and WARC headers were not represented in the JSON data in WAT files. When a header was repeated adding a further value of that header, only the last value was stored and other values were lost. This issues was fixed with CC-MAIN-2024-51, see ia-web-commons#18. All WAT files from CC-MAIN-2013-20 until CC-MAIN-2024-46 are affected.
Erratum:
Erroneous title field in WAT records
The "Title" extracted in WAT records to the JSON path `Envelope > Payload-Metadata > HTTP-Response-Metadata > HTML-Metadata > Head > Title` is not the content included in the <title> element in the HTML header (<head> element) if the page contains further <title> elements in the page body. The content of the last <title> element is written to the WAT "Title". This bug was observed if the HTML page includes embedded SVG graphics.
The issue was reported by the user Robert Waksmunski:
- https://groups.google.com/g/common-crawl/c/ZrPFdY3pPA4/m/s5D_8wCJAAAJ
- WAT extractor: Document title bug ia-web-commons#36
...and was fixed for CC-MAIN-2024-42 by commoncrawl/ia-web-commons#37.
This erratum affects all crawls from CC-MAIN-2013-20 until CC-MAIN-2024-38.
Erratum:
Missing Language Classification
Starting with crawl CC-MAIN-2018-39 we added a language classification field (‘content-languages’) to the columnar indexes, WAT files, and WARC metadata for all subsequent crawls. The CLD2 classifier was used, and includes up to three languages per document. We use the ISO-639-3 (three-character) language codes.
.svg)
