v1.12.0

lm-sys/RouteLLMv1.12.0Jul 30, 2024by adbar

AI Summary

Focuses on faster and more accurate web content extraction with stricter output format enforcement and various bug fixes.

Key Highlights

Enforce fixed list of output formats (breaking change).
Improved link and structure checks for extraction accuracy.
Fix MemoryError and ValueError during table conversion to text.
XML-TEI updated to use DTD instead of RelaxNG.

Breaking Changes

enforce fixed list of output formats, deprecate `-out` on the CLI (#647)

New Features

review link and structure checks (#653)
improve justext fallback (#652)
baseline: prevent LXML error in JSON-LD (#643), do not use as backup extraction (#646)
review XPaths for undesirable content (#645)
CLI fix: markdown format should trigger `include_formatting` (#649)
images fix: use a length threshold on src attribute (#654)
XML-TEI: replace RelaxNG by DTD, remove pickle, and update (#655)
formatting & markdown fix: add newlines (#656)
table fix: prevent `MemoryError` & `ValueError` during conversion to text (#658)

Full Release Notes

Breaking change:
- enforce fixed list of output formats, deprecate `-out` on the CLI (#647)

Faster, more accurate extraction:
- review link and structure checks (#653)
- improve justext fallback (#652)
- baseline: prevent LXML error in JSON-LD (#643), do not use as backup extraction (#646)
- review XPaths for undesirable content (#645)

Bugfixes and maintenance:
- CLI fix: markdown format should trigger `include_formatting` (#649)
- images fix: use a length threshold on src attribute (#654)
- XML-TEI: replace RelaxNG by DTD, remove pickle, and update (#655)
- formatting & markdown fix: add newlines (#656)
- table fix: prevent `MemoryError` & `ValueError` during conversion to text (#658)

Documentation:
- update `crawls.rst`: `known` is an unexpected argument, by @tommytyc in #638