Commit message (Collapse) | Author | Age | Files | Lines | ||
---|---|---|---|---|---|---|
... | ||||||
* | | another bad PDF SHA-1 | Bryan Newbold | 2020-06-30 | 1 | -0/+1 | |
| | | ||||||
* | | hack to unblock thumbnail processing pipeline | Bryan Newbold | 2020-06-29 | 1 | -0/+16 | |
| | | | | | | | | | | | | Some PDFs taking 10+ minutes to process, causing kafka exceptions and consumer churn. Not sure why kafka json pusher timeouts are not catching these. | |||||
* | | customize timeout per worker; 120sec for pdf-extract | Bryan Newbold | 2020-06-29 | 3 | -2/+4 | |
| | | | | | | | | | | This is a stab-in-the-dark attempt to resolve long timeouts with this worker in prod. | |||||
* | | handle empty fetched blob | Bryan Newbold | 2020-06-27 | 1 | -1/+6 | |
| | | ||||||
* | | CDX KeyError as WaybackError from fetch worker | Bryan Newbold | 2020-06-26 | 1 | -1/+1 | |
| | | ||||||
* | | handle None 'metadata' field correctly | Bryan Newbold | 2020-06-26 | 1 | -1/+1 | |
| | | ||||||
* | | handle non-success case of parsing extract from JSON/dict | Bryan Newbold | 2020-06-26 | 1 | -1/+1 | |
| | | ||||||
* | | report revisit non-200 as a WaybackError | Bryan Newbold | 2020-06-26 | 1 | -7/+7 | |
| | | ||||||
* | | Revert "simpler handling of null PDF text pages" | Bryan Newbold | 2020-06-25 | 1 | -4/+11 | |
| | | | | | | | | | | | | This reverts commit 254f24ad6566c9d4b5814868911b604802847b58. Attribute was actually internal to text() call, not a None page. | |||||
* | | simpler handling of null PDF text pages | Bryan Newbold | 2020-06-25 | 1 | -11/+4 | |
| | | ||||||
* | | pdfextract: attributerror with text extraction | Bryan Newbold | 2020-06-25 | 1 | -4/+12 | |
| | | ||||||
* | | catch UnicodeDecodeError in pdfextract | Bryan Newbold | 2020-06-25 | 1 | -1/+10 | |
| | | ||||||
* | | don't nest generic fetch errors under pdf_trio | Bryan Newbold | 2020-06-25 | 1 | -12/+6 | |
| | | | | | | | | This came from sloppy refactoring (and missing test coverage) | |||||
* | | pdfextract: handle too-large fulltext | Bryan Newbold | 2020-06-25 | 1 | -0/+17 | |
| | | ||||||
* | | another bad/non PDF test; catch correct error | Bryan Newbold | 2020-06-25 | 2 | -1/+6 | |
| | | | | | | | | | | | | This test doesn't actually catch the error. I'm not sure why type checks don't discover the "LockedDocumentError not part of poppler" issue though. | |||||
* | | pdfextract: catch poppler.LockedDocumentError | Bryan Newbold | 2020-06-25 | 1 | -1/+1 | |
| | | ||||||
* | | commented special modes for dump_unextracted_pdf.sql | Bryan Newbold | 2020-06-25 | 1 | -1/+4 | |
| | | ||||||
* | | pdfextract support in ingest worker | Bryan Newbold | 2020-06-25 | 3 | -1/+66 | |
| | | ||||||
* | | poppler: correct RGBA buffer endian-ness | Bryan Newbold | 2020-06-25 | 2 | -2/+2 | |
| | | ||||||
* | | 2020-05_pubmed ingest notes (short) | Bryan Newbold | 2020-06-25 | 1 | -0/+10 | |
| | | ||||||
* | | commit old notes on a one-off CDX table cleanup | Bryan Newbold | 2020-06-25 | 1 | -0/+34 | |
| | | ||||||
* | | commit old (2020-02) pdftrio commands | Bryan Newbold | 2020-06-25 | 1 | -0/+162 | |
| | | ||||||
* | | pdftrio SQL queries | Bryan Newbold | 2020-06-25 | 1 | -0/+65 | |
| | | ||||||
* | | SQL commands for re-trying PDF ingests | Bryan Newbold | 2020-06-25 | 1 | -0/+158 | |
| | | ||||||
* | | start of RUNBOOK commands | Bryan Newbold | 2020-06-25 | 1 | -0/+44 | |
| | | ||||||
* | | unextracted PDF job dump command | Bryan Newbold | 2020-06-25 | 1 | -0/+16 | |
| | | ||||||
* | | pdfextract_tool fixes from prod usage | Bryan Newbold | 2020-06-25 | 2 | -3/+6 | |
| | | ||||||
* | | fix tests for page0_height/width | Bryan Newbold | 2020-06-25 | 1 | -2/+2 | |
| | | ||||||
* | | pdfextract: fix pdf_extra key names | Bryan Newbold | 2020-06-25 | 1 | -2/+2 | |
| | | ||||||
* | | ensure pdf_meta isn't passed an empty dict() | Bryan Newbold | 2020-06-25 | 1 | -1/+4 | |
| | | ||||||
* | | args.kafka_env refactor didn't happen (yet) | Bryan Newbold | 2020-06-25 | 1 | -2/+2 | |
| | | ||||||
* | | s3-only mode persist workers use different consumer group | Bryan Newbold | 2020-06-25 | 1 | -2/+8 | |
| | | ||||||
* | | changes from prod | Bryan Newbold | 2020-06-25 | 2 | -4/+18 | |
| | | ||||||
* | | PDF extraction kafka topics | Bryan Newbold | 2020-06-25 | 1 | -0/+24 | |
| | | ||||||
* | | sandcrawler_worker: remove duplicate run_pdf_extract() | Bryan Newbold | 2020-06-25 | 1 | -29/+0 | |
| | | ||||||
* | | gitlab-ci: install most recent cmake via PPA | Bryan Newbold | 2020-06-25 | 1 | -3/+7 | |
| | | ||||||
* | | pdfextract worker | Bryan Newbold | 2020-06-25 | 1 | -1/+34 | |
| | | ||||||
* | | pdfextract: don't compress thumbnail output | Bryan Newbold | 2020-06-25 | 1 | -1/+1 | |
| | | ||||||
* | | pipenv: python-poppler 0.2.1 | Bryan Newbold | 2020-06-25 | 2 | -49/+51 | |
| | | ||||||
* | | add xenial poppler PPA (PDF parsing) | Bryan Newbold | 2020-06-18 | 1 | -0/+1 | |
| | | ||||||
* | | fixes and tweaks from testing locally | Bryan Newbold | 2020-06-17 | 6 | -18/+134 | |
| | | ||||||
* | | fixes to pdfextract_tool | Bryan Newbold | 2020-06-17 | 1 | -12/+8 | |
| | | ||||||
* | | tweak pdf_meta SQL schema | Bryan Newbold | 2020-06-17 | 2 | -5/+31 | |
| | | ||||||
* | | add a docker-compose file for local development | Bryan Newbold | 2020-06-17 | 2 | -0/+50 | |
| | | ||||||
* | | write up blob storage (w/ seaweedfs) | Bryan Newbold | 2020-06-17 | 2 | -0/+95 | |
| | | | | | | | | This follows-on to Martin's earlier proposal. | |||||
* | | tweak kafka topic names and seaweedfs layout | Bryan Newbold | 2020-06-17 | 3 | -13/+16 | |
| | | ||||||
* | | make process_pdf() more robust to parse errors | Bryan Newbold | 2020-06-17 | 1 | -5/+29 | |
| | | ||||||
* | | note about text layout with pdf extraction | Bryan Newbold | 2020-06-17 | 1 | -0/+8 | |
| | | ||||||
* | | lint fixes | Bryan Newbold | 2020-06-17 | 4 | -7/+9 | |
| | | ||||||
* | | rename pdf tools to pdfextract | Bryan Newbold | 2020-06-17 | 3 | -0/+0 | |
| | |