fetch_util 0.6.0 → 0.6.2
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- checksums.yaml +4 -4
- data/CHANGELOG.md +35 -0
- data/lib/fetch_util/assets/extract.js +1 -1
- data/lib/fetch_util/version.rb +1 -1
- metadata +1 -1
checksums.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
---
|
|
2
2
|
SHA256:
|
|
3
|
-
metadata.gz:
|
|
4
|
-
data.tar.gz:
|
|
3
|
+
metadata.gz: 48a3acb88a4254f45ac57c889570f0f70b2fcae20e42957164d7b23114f18226
|
|
4
|
+
data.tar.gz: 21525e77346c57d53ce78070654462817990bb6bbdb6a229e6e9ad283940cc0d
|
|
5
5
|
SHA512:
|
|
6
|
-
metadata.gz:
|
|
7
|
-
data.tar.gz:
|
|
6
|
+
metadata.gz: 9b511e0a3a8603b3d577595fe98870b2b2563123efb99c2afd8a2b807eda326f0d62a110d6d3262285fcac3744b0ba1647046da20c1f669ca6405d05ac5a5e21
|
|
7
|
+
data.tar.gz: 2840c6daad1e744558d6c9c7ee82ce0c873113437f7ed28ff511316322bdc0ce871de63bb0cff836607083841d930dee7af94f50292b8e7509435917162fda1b
|
data/CHANGELOG.md
CHANGED
|
@@ -2,6 +2,41 @@
|
|
|
2
2
|
|
|
3
3
|
## Unreleased
|
|
4
4
|
|
|
5
|
+
## v0.6.2 - 2026-09-10
|
|
6
|
+
|
|
7
|
+
### Changed
|
|
8
|
+
|
|
9
|
+
- Prefer generic and shared extraction over lossy or unreachable host profiles, removing eleven redundant Drupal, Rails API, Fandom, O País, France24, Dinakaran, MediaWiki, Der Standard, Top Channel, GeoTV, and Sözcü registrations while retaining profiles with demonstrated ownership or warning behavior.
|
|
10
|
+
- Register the FAZ article extractor directly while preserving its separate homepage behavior.
|
|
11
|
+
|
|
12
|
+
### Fixed
|
|
13
|
+
|
|
14
|
+
- Preserve complete visible liveblogs and owned article context, metadata, media, captions, and categories across NHK, Jang, Kurir, Gazeta, 20minutos, Zeit, Civil.ge, and Ghost pages while continuing to reject attached navigation, share controls, comments, and related feeds.
|
|
15
|
+
- Restore linked scholarly sections and complete bibliographies, substantive article sidebars and rails, and metadata-confirmed article leads with their locally attached hero figures.
|
|
16
|
+
- Improve generic language, title, and byline finalization for stale Open Graph locales, title-like opening prose, timestamps, date-only values, invalid author URLs, and author-section placeholders.
|
|
17
|
+
- Keep WordPress post templates out of the broad static-site profile and prefer owned post descriptions over earlier related-entry cards.
|
|
18
|
+
|
|
19
|
+
### Known Limitations
|
|
20
|
+
|
|
21
|
+
- An immutable-DOM comparison of 600 websites against v0.6.1 found 26 improvements, 569 ties, no regressions, four mixed outcomes, and one dual-timeout inconclusive. Mixed and inconclusive outcomes are not wins, so this does not establish a universal strict improvement.
|
|
22
|
+
|
|
23
|
+
## v0.6.1 - 2026-09-10
|
|
24
|
+
|
|
25
|
+
### Fixed
|
|
26
|
+
|
|
27
|
+
- Preserve visible homepage actions, wrapped semantic records, short paired-media cards, search tools, scholarly citation exports, quotations and record-owned fields without reviving inactive panels or navigation.
|
|
28
|
+
- Keep supporting citations, source lists, author profiles and exact fragment references with their nearest record while avoiding sibling ownership, fallback pseudo-records, duplicate primary links and repeated article prose.
|
|
29
|
+
- Preserve unrendered card summaries, broader homepage descriptions and stale-opacity record headings while rejecting nested noise collections and incomplete fallback ownership.
|
|
30
|
+
- Retain safe inline references and selected details in table rows and descriptions, including records discovered outside the primary table-index path.
|
|
31
|
+
|
|
32
|
+
### Performance
|
|
33
|
+
|
|
34
|
+
- Reuse visibility, section, headline and supporting-card ownership proofs within each immutable extraction pass, and skip unrelated product and scholarly probes without raising timeouts or capping output.
|
|
35
|
+
|
|
36
|
+
### Known Limitations
|
|
37
|
+
|
|
38
|
+
- The latest 300-site comparison against v0.5.2 has no high-severity regressions, but still contains one medium and twenty low regressions plus mixed outcomes. This patch does not establish a generalized strict win.
|
|
39
|
+
|
|
5
40
|
## v0.6.0 - 2026-09-08
|
|
6
41
|
|
|
7
42
|
### Added
|