text stringlengths 3 8.33k | repo stringclasses 52
values | path stringlengths 6 141 | language stringclasses 35
values | sha stringlengths 64 64 | chunk_index int32 0 273 | n_tokens int32 1 896 |
|---|---|---|---|---|---|---|
unit/test_scrapers/test_html_crawler.py tests/unit/test_scrapers/test_web_crawl.py tests/unit/test_config.py -v`: 92 passed
- `pytest tests/ -m 'not live' -x -q`: 636 passed, 39 deselected (live), 0 failures
- Config check: `s.source_variants['bioinf_jku'].crawl_mode == 'recursive'` — OK
- Scraper check: `get_scraper('... | jku-encyclopedia | .planning/phases/12-web-crawl-infrastructure/12-02-SUMMARY.md | Markdown | 6b7b89021d01d78de735460ea270157b85e46aca9a9a7bebf7a933d66eb1a2ac | 1 | 239 |
# Phase 12: Web Crawl Infrastructure - Context
**Gathered:** 2026-03-23
**Status:** Ready for planning
<domain>
## Phase Boundary
Recursive HTML crawler replaces sitemap-only discovery, enabling all 6 institute scrapers that have no XML sitemap. The crawler can follow anchor links within the same domain to discover ... | jku-encyclopedia | .planning/phases/12-web-crawl-infrastructure/12-CONTEXT.md | Markdown | 5bed2138166f6a7db1ee05b811d35ce0895e4c476f9b60cb0fb42d35f219d47a | 0 | 896 |
based HTML text extraction with main/article content detection
- `_extract_pdf_links_from_html()` in web_crawl.py: PDF link extraction side-effect — call during recursive crawl too
- `_url_to_id()` in web_crawl.py: URL-to-item-ID conversion with truncation
- `BaseScraper._rate_limited_get()`: rate-limited HTTP GET with... | jku-encyclopedia | .planning/phases/12-web-crawl-infrastructure/12-CONTEXT.md | Markdown | 91ca0ba498e72d06d690791c0d38d47a68ee0c2f2821cee9438b3a1a0f96102a | 1 | 300 |
# Phase 12: Web Crawl Infrastructure - Research
**Researched:** 2026-03-23
**Domain:** Async Python web crawling — BFS, robots.txt, URL normalization, SourceConfig extension
**Confidence:** HIGH
## Summary
Phase 12 adds recursive HTML crawling to the existing `WebCrawlScraper` infrastructure. All six institute sourc... | jku-encyclopedia | .planning/phases/12-web-crawl-infrastructure/12-RESEARCH.md | Markdown | bd818acda4f786a643ec6d041b6760b33048d3179b1b977637b12d5093df6d3c | 0 | 896 |
Support |
|----|-------------|-----------------|
| CRAWL-01 | Web crawler can recursively discover pages from HTML start URLs (not just XML sitemaps) | `html_crawler.py` BFS function + `crawl_mode` branching in `discover()` + `SourceConfig` extension |
</phase_requirements>
---
## Standard Stack
### Core (already in... | jku-encyclopedia | .planning/phases/12-web-crawl-infrastructure/12-RESEARCH.md | Markdown | 9dd929fff2d1578f21615aba70433dced3e2c98e547a1db13b7f8bc575672959 | 1 | 896 |
`WebCrawlScraper.discover()` passes a lambda wrapping `self._rate_limited_get`, so the per-domain `AsyncLimiter` already in `BaseScraper` is reused automatically — no new rate-limiting code needed in the crawler.
### Pattern 2: crawl_mode Branching in discover()
**What:** `WebCrawlScraper.discover()` switches on `sel... | jku-encyclopedia | .planning/phases/12-web-crawl-infrastructure/12-RESEARCH.md | Markdown | e67b61994da17a1a18a10e347553018606fde7c5a60658f756c6b0fa139f1259 | 2 | 896 |
recursive'
js_rendering: bool = False
```
Then update all five institute `source_variants` in `Settings` to include `crawl_mode='recursive'`, and the `jku_web` scraper (registered via `@register_scraper`) uses `crawl_mode='hybrid'` (or receives it from its SourceConfig if refactored, but jku_web is currently class... | jku-encyclopedia | .planning/phases/12-web-crawl-infrastructure/12-RESEARCH.md | Markdown | 1646a53cb63509ecefc4d87b312106b27b34e19402a5843ba0c20bc3852ff82c | 3 | 896 |
crawl-delay correctly; stdlib, zero install cost |
| URL joining | String concatenation | `urllib.parse.urljoin(base, href)` (stdlib) | Handles relative, absolute, protocol-relative, and path-relative hrefs correctly |
| URL normalization | Custom regex | `urllib.parse.urlparse` + `._replace()` | Fragment stripping and... | jku-encyclopedia | .planning/phases/12-web-crawl-infrastructure/12-RESEARCH.md | Markdown | e87cca4f3e99d563d0116a65afcbcae56a438ab8a69294ad153f80c808da04dd | 4 | 896 |
has the hardcoded `if not parsed.netloc.endswith(".jku.at"): continue` guard. Institute domains ARE `*.jku.at` subdomains, so this is fine — but the guard must not change the behavior for institute scrapers.
**Why it happens:** The existing guard in `WebCrawlScraper.discover()` (lines 87-91 in `web_crawl.py`) is a seco... | jku-encyclopedia | .planning/phases/12-web-crawl-infrastructure/12-RESEARCH.md | Markdown | 4f0da299fe856040fb6b90d5a2486bace6ef7b63a04f044badf1e9440d6d63b3 | 5 | 896 |
.Response]],
url_passes_filter: Callable[[str], bool],
max_pages: int = 500,
) -> list[str]:
"""BFS crawl from start_url; return list of same-domain, filter-passing URLs."""
start_domain = urlparse(start_url).netloc.lower()
visited: set[str] = set()
queue: deque[str] = deque()
robots_cache: ... | jku-encyclopedia | .planning/phases/12-web-crawl-infrastructure/12-RESEARCH.md | Markdown | b1e437d8d5a636fe7752842899361aba22810299a861af51c88ca85e1b49a989 | 6 | 896 |
update(discovered)
# else: unknown crawl_mode — log warning
items: list[RawItem] = []
rejected = 0
for url in sorted(urls)[: self.max_pages]:
parsed = urlparse(url)
if not parsed.netloc.endswith(".jku.at"):
continue
if not _url_passes_filter(url, self.include_pat... | jku-encyclopedia | .planning/phases/12-web-crawl-infrastructure/12-RESEARCH.md | Markdown | deb10f67111a31626b0d9349f4073222c49ab45f295c79fe726902b360effdbf | 7 | 896 |
test_scrapers/test_html_crawler.py tests/unit/test_scrapers/test_web_crawl.py tests/unit/test_config.py -x` |
| Full suite command | `pytest tests/ -m 'not live'` |
### Phase Requirements → Test Map
| Req ID | Behavior | Test Type | Automated Command | File Exists? |
|--------|----------|-----------|-----------------... | jku-encyclopedia | .planning/phases/12-web-crawl-infrastructure/12-RESEARCH.md | Markdown | 36ea4247581e161bd707ef362d97311b1e7f258cede9d961379c2eb5e28640f6 | 8 | 896 |
:**
- Standard stack: HIGH — all libraries already in pyproject.toml; stdlib confirmed available
- Architecture: HIGH — based on direct code reading; patterns extend existing established patterns
- Pitfalls: HIGH — derived from reading the existing code and standard BFS crawling failure modes
- URL normalization: HIGH ... | jku-encyclopedia | .planning/phases/12-web-crawl-infrastructure/12-RESEARCH.md | Markdown | c20d8ab85eabc171b46a2b35eb30d7344175653bd9b0bc4088e3e2e2594dbbd2 | 9 | 104 |
---
phase: 12
slug: web-crawl-infrastructure
status: draft
nyquist_compliant: false
wave_0_complete: false
created: 2026-03-23
---
# Phase 12 — Validation Strategy
> Per-phase validation contract for feedback sampling during execution.
---
## Test Infrastructure
| Property | Value |
|----------|-------|
| **Framew... | jku-encyclopedia | .planning/phases/12-web-crawl-infrastructure/12-VALIDATION.md | Markdown | 9f8ce9a5aedc074c4ff7d6d966e49ec67c9af95a1fb278ec1672cfac9c7275fe | 0 | 760 |
---
phase: 12-web-crawl-infrastructure
verified: 2026-03-23T00:00:00Z
status: passed
score: 15/15 must-haves verified
re_verification: false
---
# Phase 12: Web Crawl Infrastructure Verification Report
**Phase Goal:** The web crawler can recursively discover pages from any HTML start URL, not just XML sitemaps — enab... | jku-encyclopedia | .planning/phases/12-web-crawl-infrastructure/12-VERIFICATION.md | Markdown | 89c868b1586bb76ec8e8a25a11faeb71655829004fb5b0e72995a0ece8889c26 | 0 | 896 |
-------|
| `src/jku_kb/scrapers/html_crawler.py` | Standalone async BFS crawl function | VERIFIED | 221 lines; exports `crawl_html`, `_normalize_url`, `_extract_links`; contains `RobotFileParser`, `deque`, `BeautifulSoup`, `SKIP_EXTENSIONS`, `USER_AGENT` |
| `tests/unit/test_scrapers/test_html_crawler.py` | Unit tests ... | jku-encyclopedia | .planning/phases/12-web-crawl-infrastructure/12-VERIFICATION.md | Markdown | 8f79167953056b25dc515dd63b6ddd2a26a2e85ad6bacbfb41c6697fb7c8bf37 | 1 | 896 |
Phase 13.
---
### Anti-Patterns Found
No anti-patterns found across all phase files.
Note: `return []` at html_crawler.py line 66 is a legitimate error-handling fallback (empty link list when BeautifulSoup fails to parse HTML), not a stub.
---
### Human Verification Required
None. All phase-12 behaviors are full... | jku-encyclopedia | .planning/phases/12-web-crawl-infrastructure/12-VERIFICATION.md | Markdown | 1fdd12225db77d49d5486a597f0f05f9a9531c8f806ef5fc490e77bf04f66112 | 2 | 462 |
---
phase: 13-institute-web-crawlers
plan: 01
type: execute
wave: 1
depends_on: []
files_modified:
- src/jku_kb/scrapers/web_crawl.py
- src/jku_kb/scrapers/__init__.py
- src/jku_kb/config.py
- tests/unit/test_scrapers/test_web_crawl.py
autonomous: true
requirements: [CRAWL-02]
must_haves:
truths:
- "DEFA... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-01-PLAN.md | Markdown | d8c3e102f19ad868283d1b0afc2b53590a29317bd32f61be062bb23e4dcec3d9 | 0 | 896 |
['sitemap', 'recursive', 'hybrid'] = 'recursive'
js_rendering: bool = False
```
</interfaces>
</context>
<tasks>
<task type="auto" tdd="true">
<name>Task 1: Add DEFAULT_EXCLUDE_PATTERNS and merge in discover()</name>
<files>src/jku_kb/scrapers/web_crawl.py, tests/unit/test_scrapers/test_web_crawl.py</files>
... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-01-PLAN.md | Markdown | eb26ecda02224b441a03bc329548dc85df14d0f448a0e4037a8754276619f0ea | 1 | 896 |
tests/unit/test_scrapers/test_web_crawl.py -x -q` exits 0 (all existing tests still pass)
</acceptance_criteria>
<done>DEFAULT_EXCLUDE_PATTERNS constant exists in web_crawl.py, merged into all discover() filter paths, 9 new tests pass, all existing tests pass</done>
</task>
<task type="auto" tdd="true">
<name>Ta... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-01-PLAN.md | Markdown | 7d36e09a6dbe3158c50c2b4e69b3f935225ca0181694bfdb899fbfa81285db37 | 2 | 896 |
/test_web_crawl.py -k "TestJkuWebHybridConfig" -x -q && pytest tests/unit/test_scrapers/test_web_crawl.py -x -q</automated>
</verify>
<acceptance_criteria>
- config.py contains `"jku_web": SourceConfig(` inside source_variants dict
- config.py jku_web entry has `crawl_mode="hybrid"`
- config.py jku_web ... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-01-PLAN.md | Markdown | d10bc09d7cb24c8af528d66b74308c7d810757a54b941d282b52de61017b27f2 | 3 | 407 |
---
phase: 13-institute-web-crawlers
plan: 01
subsystem: scraping
tags: [web-crawl, url-filtering, hybrid-mode, source-variants, pydantic]
# Dependency graph
requires:
- phase: 12-web-crawl-infrastructure
provides: WebCrawlScraper with crawl_mode support and crawl_html function
provides:
- DEFAULT_EXCLUDE_PAT... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-01-SUMMARY.md | Markdown | 381b89d91458054601776318dac700f4f1166711e32780709f08c4d6e0f7a580 | 0 | 896 |
.xml and HTML entry URLs so hybrid mode handles full jku.at coverage
## Deviations from Plan
None - plan executed exactly as written.
## Issues Encountered
None
## User Setup Required
None - no external service configuration required.
## Next Phase Readiness
- DEFAULT_EXCLUDE_PATTERNS ready for all institute crawl... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-01-SUMMARY.md | Markdown | 39b469f555421e101d39b57c5b05e08083851a43a1c072405c4907af46e26b35 | 1 | 214 |
---
phase: 13-institute-web-crawlers
plan: 02
type: execute
wave: 2
depends_on: ["13-01"]
files_modified:
- src/jku_kb/config.py
- tests/unit/test_scrapers/test_web_crawl.py
autonomous: true
requirements: [CRAWL-03, CRAWL-04, CRAWL-05]
must_haves:
truths:
- "bioinf_jku scraper discovers pages using broad sta... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-02-PLAN.md | Markdown | 21b379cda770d290561d5e6f8c53a1480fbfa8a9ad26729d410b279167ee74f0 | 0 | 896 |
"/team/"],
),
"cp_jku": SourceConfig(
scraper_class="web_crawl",
start_urls=["https://www.cp.jku.at/teaching/"],
max_pages=500,
include_patterns=["/teaching/", "/research/", "/publications/", "/datasets/"],
exclude_patterns=["/contact", "/imprint", "/news/"],
),
```
From src/jku_kb/scrapers/__init_... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-02-PLAN.md | Markdown | 491bd044e7bb9993b08e1980abdfc025e1efa68385e1aee2e4965d36048b62bc | 1 | 896 |
python
"cp_jku": SourceConfig(
scraper_class="web_crawl",
start_urls=[
"https://www.cp.jku.at/",
"https://www.cp.jku.at/teaching/",
"https://www.cp.jku.at/research/",
"https://www.cp.jku.at/people/",
],
max_pages=500,
include_patterns=[],
exclude_patterns=[],
),
`... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-02-PLAN.md | Markdown | d9cb8cb7bed12fb80dfdcde60f1537518ad2814caa8860277df902366c85d49f | 2 | 748 |
---
phase: 13-institute-web-crawlers
plan: 02
subsystem: scraping
tags: [web-crawl, source-variants, config, institute-crawlers, bioinf, fmv, cp]
# Dependency graph
requires:
- phase: 13-institute-web-crawlers
plan: 01
provides: DEFAULT_EXCLUDE_PATTERNS merged in discover(), source_variants precedence in get... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-02-SUMMARY.md | Markdown | 5c0994afdbf7300d12a0c9869fe18ce51717d662500aa5f17e2717f41035d5e3 | 0 | 896 |
not introduced by this plan, logged to deferred-items.md
## Known Stubs
None — all 3 institute configs are fully wired with real start_urls pointing to live institute websites.
## Next Phase Readiness
- bioinf_jku, fmv_jku, cp_jku now configured for full site discovery via recursive crawl
- Plan 13-03 (ssw_jku, ris... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-02-SUMMARY.md | Markdown | d1966f6444466bc9a01dd59c9da5636a537e84bdd18e62e0d58b8c6950216d02 | 1 | 142 |
---
phase: 13-institute-web-crawlers
plan: 03
type: execute
wave: 2
depends_on: ["13-01"]
files_modified:
- src/jku_kb/config.py
- data/sources.json
- tests/unit/test_scrapers/test_web_crawl.py
autonomous: true
requirements: [CRAWL-06, CRAWL-07]
must_haves:
truths:
- "ssw_jku scraper discovers pages using ... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-03-PLAN.md | Markdown | 9e21ea166066073a546bc6b4ebed16d47831bb47b5dc7ea50e698c66ea467efe | 0 | 896 |
//www3.risc.jku.at/education/courses/",
],
max_pages=1000,
include_patterns=["/publications/", "/education/courses/", "/research/"],
exclude_patterns=["/people/contact", "/about/imprint"],
),
```
From data/sources.json (existing institute entries use old source_ids):
- "bioinf_teaching" (not "bioinf_jk... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-03-PLAN.md | Markdown | dff08268878e34d4b367bd1a38d40c8f3524db73d9f12ba06e5cb0615e3d6b51 | 1 | 896 |
preserved as start_url for direct publication listing access.
3. **Add TestSswRiscConfigs class** in test_web_crawl.py with 8 tests. For test 4 (ssw discover), mock crawl_html to return capital-case URLs and verify they pass through discover() correctly:
```python
@pytest.mark.asyncio
async def test_ssw_jku_discover_... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-03-PLAN.md | Markdown | d7d600d522d7e242b32460936a177428dda11b118640a2aa13067a7f195004d1 | 2 | 896 |
/www.bioinf.jku.at/",
"listing_mechanism": "crawl",
"auth_required": false,
"embedding_modality": "text",
"chunking_strategy": "token_windows",
"language": "en",
"license": "copyright",
"robots_txt_compliant": true,
"rate_limit_delay_seconds": 2,
"requires_javascript": false,
"priority": "high",
"... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-03-PLAN.md | Markdown | 488ded55ea564285b885477e339079de10dec40d38bd441f51a89ef1f1f3098b | 3 | 896 |
json entries exist:
```python
class TestSourcesJsonEntries:
"""Verify sources.json has quality gate entries for all 6 institute scrapers."""
def test_institute_source_ids_in_sources_json(self):
import json
with open("data/sources.json") as f:
data = json.load(f)
source_ids ... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-03-PLAN.md | Markdown | ce9ea145a2a697fbd17f9858403609412ad8f465a936bb6b241de2cc9102ffa7 | 4 | 714 |
---
phase: 13-institute-web-crawlers
plan: 03
subsystem: scraping
tags: [web_crawl, config, sources_json, quality_gate, ssw_jku, risc_web, institute]
# Dependency graph
requires:
- phase: 13-01
provides: DEFAULT_EXCLUDE_PATTERNS merged in discover(), source_variants support in get_scraper()
provides:
- ssw_jku... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-03-SUMMARY.md | Markdown | abca7914df9b5d69d04050ab233939df59e310e58426a760d8de2386d9a6d149 | 0 | 896 |
Decisions Made
- ssw_jku uses capital-case start_urls (/Teaching/, /Research/) with include_patterns=[] to avoid case-sensitivity pitfall — no regex matching on path case needed. Since include_patterns is empty, _url_passes_filter returns True for all URLs (exclude-only mode).
- risc_web PHP publications URL preserved ... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-03-SUMMARY.md | Markdown | e591597f8c9c09d2d0ce5310b6e550a3aef55655add1595c731fa1af294481f3 | 1 | 392 |
# Phase 13: Institute Web Crawlers - Context
**Gathered:** 2026-03-23
**Status:** Ready for planning
<domain>
## Phase Boundary
Wire up all 6 institute web scrapers (jku_web, bioinf_jku, fmv_jku, cp_jku, ssw_jku, risc_web) to use the recursive HTML crawler from Phase 12, tune each scraper's configuration against its... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-CONTEXT.md | Markdown | 9a1c527747a980afe413115c7df8ea6c6a0d5403240af20b5ec0707de7d6ecdd | 0 | 896 |
</canonical_refs>
<code_context>
## Existing Code Insights
### Reusable Assets
- `crawl_html()` in html_crawler.py: fully built BFS crawler accepting fetch callable and url_passes_filter predicate — ready to use
- `_url_passes_filter()` in web_crawl.py: regex-based include/exclude filtering — already wired into disco... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-CONTEXT.md | Markdown | 187dae45195a0023d1084a25c813f2a4cd36807877ec26bb03b3dc98b409e831 | 1 | 308 |
---
status: partial
phase: 13-institute-web-crawlers
source: [13-VERIFICATION.md]
started: 2026-03-23T00:00:00Z
updated: 2026-03-23T00:00:00Z
---
## Current Test
[awaiting human testing]
## Tests
### 1. Live smoke run — each of the 6 scrapers against their live endpoints
expected: bioinf_jku/fmv_jku/cp_jku/ssw_jku/... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-HUMAN-UAT.md | Markdown | 48fa9e8c59eec90a389d4d9d517bd111a2eea64dfdc93b1832509d46aff69a45 | 0 | 136 |
# Phase 13: Institute Web Crawlers - Research
**Researched:** 2026-03-23
**Domain:** Web scraper configuration and tuning — wiring 6 institute scrapers to the Phase 12 BFS HTML crawler
**Confidence:** HIGH
---
<user_constraints>
## User Constraints (from CONTEXT.md)
### Locked Decisions
**Start URL strategy**
- Mu... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-RESEARCH.md | Markdown | 49cd4c30f20ad9304acc661063add3c543f07c64c203a342231ac4769e457508 | 0 | 896 |
---
## Summary
Phase 13 is a configuration and tuning phase — no new crawler infrastructure is required. The BFS crawler (`crawl_html`) and the `WebCrawlScraper` with `crawl_mode` support are fully built by Phase 12. The work is: (1) add `DEFAULT_EXCLUDE_PATTERNS` constant to `web_crawl.py`, (2) update each institute... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-RESEARCH.md | Markdown | 0b0169fabda3789d5403034282e39ccd47f56f3a0203c38439aa6e842da55404 | 1 | 896 |
hybrid"` and `max_pages=50000`
Option 2 is the minimal change — update the defaults in `WebCrawlScraper.__init__` for `jku_web`. However, since the class also serves as the base for source_variants, the planner should note that changing defaults on the class affects all callers. The safer approach is to keep `crawl_mo... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-RESEARCH.md | Markdown | ea2eb9cce2b465f4533e29da352b42bb61837cdce86ca0c2b9515ba13a51a171 | 2 | 896 |
rate limiting | Custom retry logic | `BaseScraper._rate_limited_get` | Already wired to `crawl_html` via WebCrawlScraper |
| Source quality gates | Custom checker | `SourceQuality.expected_min_items` + `_compute_health_status()` | Already in cli.py |
---
## Common Pitfalls
### Pitfall 1: include_patterns Too Narrow
... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-RESEARCH.md | Markdown | 4777afc978ffeae085142f39330a61b6fccc766776a6e66a90eee517e8163682 | 3 | 896 |
-page table) or paginated. If single-page HTML listing, max_pages=1 is sufficient and the page HTML itself is the item.
### Pitfall 6: Missing quality entry in sources.json for institute scrapers
**What goes wrong:** Source entries like `bioinf_teaching`, `fmv_teaching`, etc. exist in sources.json but do NOT have `qua... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-RESEARCH.md | Markdown | 167e9297cf078e97b36b0756e71320101a710220b623858017fed81ededa3505 | 4 | 896 |
>Research</a>
<a href="/imprint">Imprint</a>
</body></html>"""
@pytest.mark.asyncio
async def test_bioinf_jku_config_wired_correctly(settings):
"""bioinf_jku SourceConfig wires start_urls to crawl_html."""
mock_crawl = AsyncMock(return_value=[
"https://www.bioinf.jku.at/teaching/ml/",
"https:... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-RESEARCH.md | Markdown | f7edc1428d90525584aec4e8fc63a7b133268114576b4dde7917ec7711b55a53 | 5 | 896 |
count:** 50-200 pages
### risc_web (CRAWL-07)
- **Current state:** 2 start_urls (PHP publications page + education/courses/), max_pages=1000
- **Target state:** Verify PHP page structure — is it a single HTML listing or paginated? Add root if needed.
- **Site notes:** `www3.risc.jku.at` — RISC institute. PHP-driven si... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-RESEARCH.md | Markdown | efc3294c731b7ab3cac2e0515b4ed70f4091582f57f02a4c1304dae4a7c93a5d | 6 | 896 |
TestInstituteConfigs` covering CRAWL-03 through CRAWL-07 (one test per institute: config wired, crawl_html called with correct args, DEFAULT_EXCLUDE_PATTERNS merged)
- [ ] `tests/unit/test_scrapers/test_web_crawl.py` — add test `TestDefaultExcludePatterns` verifying imprint/datenschutz/contact URLs are blocked even wit... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-RESEARCH.md | Markdown | 92f977dfcb6e80f8cc9849cf64ccba5900f21a170edf23095a07f3fea2550db0 | 7 | 816 |
---
phase: 13
slug: institute-web-crawlers
status: draft
nyquist_compliant: false
wave_0_complete: false
created: 2026-03-23
---
# Phase 13 — Validation Strategy
> Per-phase validation contract for feedback sampling during execution.
---
## Test Infrastructure
| Property | Value |
|----------|-------|
| **Framewor... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-VALIDATION.md | Markdown | 87a3fb7d9b987fbb26c7cd35958b4a9d6343dac015f12c82317b9ffe120a09f6 | 0 | 896 |
-----|-------------|------------|-------------------|
| jku_web discovers 1000+ items from live jku.at | CRAWL-02 | Live site dependency | Run `jku-kb scrape jku_web --dry-run` and verify count >= 1000 |
| bioinf_jku discovers teaching/research/publications | CRAWL-03 | Live site dependency | Run `jku-kb scrape bioinf_... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-VALIDATION.md | Markdown | 6ce2b5be326895fefe277cc1200700ba592f1f83c131ff2d4587192e221a44e4 | 1 | 348 |
---
phase: 13-institute-web-crawlers
verified: 2026-03-23T00:00:00Z
status: passed
score: 9/9 must-haves verified
re_verification: false
human_verification:
- test: "Live smoke run — each of the 6 scrapers against their live endpoints"
expected: "bioinf_jku/fmv_jku/cp_jku/ssw_jku/risc_web each produce >= 20 items... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-VERIFICATION.md | Markdown | ceacfbdfaf60fa19734fdcb5f8485bbdf9e76892e4c921040d2d8459eaf04cfd | 0 | 896 |
/web_crawl.py` | 13-01 | VERIFIED | Contains `DEFAULT_EXCLUDE_PATTERNS` constant; `effective_excludes` used in recursive branch (line 102), hybrid branch (line 122), and post-filter loop (line 140) |
| `src/jku_kb/config.py` | 13-01, 13-02, 13-03 | VERIFIED | Contains all 6 source_variants: jku_web (hybrid/50000), bioi... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-VERIFICATION.md | Markdown | 344fb7d3f4122551905c4ceed6b5468b751b7f6ee7eca07e4f939682d199ec19 | 1 | 896 |
--------|--------|
| All 32 new phase-13 tests pass | `pytest tests/ -k "TestDefaultExcludePatterns or TestJkuWebHybridConfig or TestInstituteConfigs or TestSswRiscConfigs or TestSourcesJsonEntries" -q` | 32 passed, 43 deselected in 0.14s | PASS |
| Full test suite green (excluding pre-existing Windows-path failure) | ... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-VERIFICATION.md | Markdown | 0d669d4758913a59b6666f76e6e5733af6be15043ddbfba05337d1ff046be271 | 2 | 896 |
`jku_web` — expects >= 100 items (CRAWL-02 target is 1000+)
- `bioinf_jku` — expects >= 20 items
- `fmv_jku` — expects >= 20 items
- `cp_jku` — expects >= 20 items
- `ssw_jku` — expects >= 20 items
- `risc_web` — expects >= 20 items
**Expected:** All 6 scrapers return non-zero item counts; jku_web returns 1000+ to ful... | jku-encyclopedia | .planning/phases/13-institute-web-crawlers/13-VERIFICATION.md | Markdown | 17b835d7b262f49aa238f7d7233105e884f11b17e67c5b4dd97ae8f8592e17f9 | 3 | 454 |
---
phase: 14-api-scraper-fixes
plan: 01
type: execute
wave: 1
depends_on: []
files_modified:
- src/jku_kb/scrapers/pdf_fetch.py
- tests/unit/test_scrapers/test_pdf_fetch.py
- data/sources.json
- src/jku_kb/scrapers/opencast.py
- tests/unit/test_scrapers/test_opencast.py
autonomous: true
requirements: [SRC-01... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-01-PLAN.md | Markdown | 74517ce4fb149b48fdc0c35a973f88d25acc5f34740260f05b3391b5b7877217 | 0 | 896 |
/files>
<read_first>
- src/jku_kb/scrapers/pdf_fetch.py (current _discover_risc_reports implementation)
- tests/unit/test_scrapers/test_pdf_fetch.py (current test fixtures and assertions)
- data/sources.json (find risc_technical_reports entry, note its full structure)
- src/jku_kb/scrapers/base.py (Ba... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-01-PLAN.md | Markdown | a7095c05d4a701d525dbd68c71d8381cf59e9198fb6381bdb23d9675f9d056bb | 1 | 896 |
unit/test_scrapers/test_pdf_fetch.py -x -q` exits 0
</acceptance_criteria>
<done>RISC scraper discovers PDFs from both divisions, matches any .pdf link, extracts metadata from table rows, deduplicates across divisions, and sources.json uses risc_reports source_id.</done>
</task>
<task type="auto" tdd="true">
<na... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-01-PLAN.md | Markdown | 13084bdc488e83a4f76ae51d43e9173569a12261da106cd760387418d62d5531 | 2 | 896 |
/done>
</task>
</tasks>
<verification>
- `python -m pytest tests/unit/test_scrapers/test_pdf_fetch.py tests/unit/test_scrapers/test_opencast.py -x -q` -- both test suites pass
- `python -c "import json; d=json.load(open('data/sources.json')); ids=[s['source_id'] for s in d['sources']]; assert 'risc_reports' in ids; a... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-01-PLAN.md | Markdown | 0227efb1f4802a15e909ff255470aaa762781b58d35311053f169642d43ebb9f | 3 | 252 |
---
phase: 14-api-scraper-fixes
plan: "01"
subsystem: scrapers
tags: [risc-reports, opencast, pagination, pdf-discovery, tdd]
dependency_graph:
requires: []
provides: [risc-multi-division-discovery, opencast-total-pagination]
affects: [data/sources.json, scraper-registry]
tech_stack:
added: []
patterns: [mult... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-01-SUMMARY.md | Markdown | e6951b0bd95082e2fd703fa6ff6f2ddf4fe494a3e259946cbdc59dbaa32b5426 | 0 | 896 |
and None/empty values
- Log `discovery_total` on first page and `discovery_complete_total_reached` when cap reached
- Added total-based break `if total > 0 and offset >= total` after incrementing offset
- Fallback to existing empty-result termination preserved when total is absent
## Verification Results
```
python -... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-01-SUMMARY.md | Markdown | 630df520ae5d7608a6ad46abea2ea476cea439b3fea302cee4db069d9bc991ad | 1 | 314 |
---
phase: 14-api-scraper-fixes
plan: 02
type: execute
wave: 1
depends_on: []
files_modified:
- src/jku_kb/scrapers/openalex_xref.py
- src/jku_kb/scrapers/openalex.py
- tests/unit/test_scrapers/test_openalex_xref.py
autonomous: true
requirements: [SRC-03, SRC-04]
must_haves:
truths:
- "OpenAlex scraper wri... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-02-PLAN.md | Markdown | a0153112246ff1488e21257a7712f7b9f1d4f364cc34640451a51c53d5a3bd94 | 0 | 896 |
://arxiv.org/abs/2405.12345"
# We must parse these to extract arXiv IDs.
#
# DOIs are available from work["doi"] (already selected).
```
From src/jku_kb/scrapers/base.py:
```python
class BaseScraper:
@property
def cache_dir(self) -> Path:
"""Per-source cache directory: {settings.cache_dir}/{source_id}/... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-02-PLAN.md | Markdown | 4a0b5efb2fdac2c500bd6affd4ecb5d71d99e4f80e549b74fbe4eda3f59c774e | 1 | 896 |
len("https://doi.org/"):]
return raw_doi
```
2. Modify `src/jku_kb/scrapers/openalex.py`:
- Add `import orjson` at top (alongside existing imports)
- Add `"locations"` to the `select` parameter string in discover() so we can extract arXiv landing page URLs. Change the select string to:
```python
"s... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-02-PLAN.md | Markdown | 0e1f1c14356a21d40ffc4495ac68e85d6c8596283e19bd9e7e87ca508c8cb28d | 2 | 896 |
empty string
- Test: _normalize_arxiv_id handles old-style IDs like "hep-ph/9901234v1"
- Test: _normalize_doi strips https://doi.org/ prefix
- Test: read_openalex_xref returns ([], []) when file does not exist
- Test: read_openalex_xref reads valid JSONL and returns correct lists
- Test: read_openal... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-02-PLAN.md | Markdown | 971ea139b764ea8ceb7a29d47d60a8bd77c7c3c3d899b32737c5794dbefce5b0 | 3 | 896 |
contains tests for _normalize_arxiv_id, _normalize_doi, read_openalex_xref
- File contains test_discover_writes_xref_file integration test
- `python -m pytest tests/unit/test_scrapers/test_openalex_xref.py -x -q` exits 0
- `python -m pytest tests/unit/test_scrapers/test_openalex.py -x -q` exits 0 (existing ... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-02-PLAN.md | Markdown | 19d622db45818384e0a12614705a84855d9110bdd2f267ae27dec3e88ce10136 | 4 | 287 |
---
phase: 14-api-scraper-fixes
plan: 02
subsystem: api
tags: [openalex, xref, cross-reference, arxiv, doi, jsonl, scraper]
# Dependency graph
requires:
- phase: 14-api-scraper-fixes
provides: "Phase context and research for API scraper fixes"
provides:
- "openalex_xref.py shared utility with read_openalex_xre... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-02-SUMMARY.md | Markdown | 0a55807ea0188dd9bc0ad067c450886119a24933916547521d04970b4db5e3d5 | 0 | 896 |
in the per-source cache dir (`openalex_jku/discover_xref.jsonl`), but `read_openalex_xref()` accepts the project-level `settings.cache_dir` to naturally locate it.
- arXiv IDs must be extracted from `locations[].landing_page_url` because the OpenAlex `ids` field does not include arXiv IDs.
- Normalization helpers live ... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-02-SUMMARY.md | Markdown | c22c6b27f3a83a77fdd6847926b9427de88defe1fc229b237e65094d407fcdbb | 1 | 315 |
---
phase: 14-api-scraper-fixes
plan: 03
type: execute
wave: 2
depends_on: [14-02]
files_modified:
- src/jku_kb/scrapers/arxiv.py
- src/jku_kb/scrapers/semantic_scholar.py
- tests/unit/test_scrapers/test_arxiv.py
- tests/unit/test_scrapers/test_semantic_scholar.py
autonomous: true
requirements: [SRC-03, SRC-04]... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-03-PLAN.md | Markdown | ad3ad8876d4584cd48670f2093f8c2a6feabbd21fd2f0d76eb6ba9e2789fb993 | 0 | 896 |
list[str]]:
"""Returns (arxiv_ids, dois). Empty lists if file missing."""
def _normalize_arxiv_id(raw_id: str) -> str:
"""Strip URL prefix and version suffix from arXiv ID."""
```
<!-- From arxiv.py - existing patterns -->
```python
API_BASE = "http://export.arxiv.org/api/query"
_ARXIV_QUERIES = [
'all:"J... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-03-PLAN.md | Markdown | c3df52262f1b8f6fa70a199eca63bb4b2560c05368e76d9a6d561098fd95345e | 1 | 896 |
.log.info("xref_arxiv_ids_found", count=len(arxiv_ids))
xref_items = await self._discover_by_ids(arxiv_ids, seen_ids)
items.extend(xref_items)
self.log.info("xref_discovery_complete", xref_items=len(xref_items))
# Phase 2: Text query fallback (existing logic)
for query in... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-03-PLAN.md | Markdown | f0c13867ecbba7ca124ed841fe4f8c42c14452c27754f9bc4ddca1b2cf075eff | 2 | 896 |
/acceptance_criteria>
<done>arXiv scraper uses OpenAlex cross-reference IDs via id_list API as primary discovery, keeps text queries as fallback, deduplicates across both, and normalizes arXiv IDs consistently.</done>
</task>
<task type="auto" tdd="true">
<name>Task 2: Add cross-reference discovery to Semantic Sch... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-03-PLAN.md | Markdown | 3f364e36ff1160f577d789fc77eb7436ae01a0451e0253cb19d8fce0689f8de0 | 3 | 896 |
't need to because there was only one discovery path).
- Add new method `_discover_by_dois`:
```python
async def _discover_by_dois(self, dois: list[str], seen_ids: set[str]) -> list[RawItem]:
"""Fetch papers by DOI using S2 /paper/batch endpoint."""
items: list[RawItem] = []
for... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-03-PLAN.md | Markdown | 460b3d57962a43f593ef087904f098014d7eb9cec20bba517757607c111e2ae6 | 4 | 896 |
test_semantic_scholar.py contains `test_discover_with_xref_dois`
- test_semantic_scholar.py contains `test_discover_xref_graceful_degradation`
- test_semantic_scholar.py contains `test_discover_xref_null_in_batch`
- `python -m pytest tests/unit/test_scrapers/test_semantic_scholar.py -x -q` exits 0
</accep... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-03-PLAN.md | Markdown | c090312c0f2b7ff3dc14bb0b06522deb226bdfc754a761e9aa58313f25001ba6 | 5 | 324 |
---
phase: 14-api-scraper-fixes
plan: "03"
subsystem: scrapers
tags: [arxiv, semantic-scholar, cross-reference, openalex, discovery, tdd]
dependency_graph:
requires: [14-02]
provides: [arxiv-xref-discovery, s2-xref-discovery]
affects: [arxiv_jku scraper, semantic_scholar_jku scraper]
tech_stack:
added: []
pat... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-03-SUMMARY.md | Markdown | 1ad72837b004b66d87a404f8b90831a127ce67e5966c9b493e23d0923ada1ea9 | 0 | 896 |
, null handling, API key header).
## Deviations from Plan
None - plan executed exactly as written.
## Verification
All acceptance criteria met:
- `arxiv.py` contains `from jku_kb.scrapers.openalex_xref import read_openalex_xref`
- `arxiv.py` contains `async def _discover_by_ids(self, arxiv_ids`
- `arxiv.py` contai... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-03-SUMMARY.md | Markdown | 37a29035dba3a3cbd15f55d6749af291498a4a86bc369f42d8aee28a6d4dd9fe | 1 | 365 |
# Phase 14: API Scraper Fixes - Context
**Gathered:** 2026-03-23
**Status:** Ready for planning
<domain>
## Phase Boundary
Fix four broken or massively underperforming API-based scrapers so each yields item counts at or above their expected minimums: risc_reports (0 → 50+), OpenCast (19 → 2000+), arXiv (15 → 500+), ... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-CONTEXT.md | Markdown | 23cff1f97c146324b55e9ee9b72c321354a76a2cfa36161e333470eb1eb82369 | 0 | 896 |
12), `arxiv_jku` (line 892), `semantic_scholar_jku` (line 837)
### Base infrastructure
- `src/jku_kb/scrapers/base.py` — BaseScraper template method pattern, `_rate_limited_get()`, error handling
- `src/jku_kb/scrapers/__init__.py` — Scraper registry with `register_scraper` decorator
### Existing tests
- `tests/unit/... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-CONTEXT.md | Markdown | eb4bb37211047388fde983f026a26370b9c023086d4e4946c6fda0d3dfe7d743 | 1 | 435 |
# Phase 14: API Scraper Fixes - Research
**Researched:** 2026-03-23
**Domain:** Python async scraping — HTML parsing, paginated REST APIs, cross-source referencing, JSONL manifest reading
**Confidence:** HIGH
---
<user_constraints>
## User Constraints (from CONTEXT.md)
### Locked Decisions
**RISC reports discovery... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-RESEARCH.md | Markdown | d8ba5d1b69b9879759275905b3b13cbc6504249b90cf02a9ade0c712a50213f4 | 0 | 896 |
comprehensively (expected 5000+) | Cross-reference from OpenAlex manifest + existing bulk search |
</phase_requirements>
---
## Summary
Four scrapers are broken or massively underperforming. The root causes are distinct for each:
**RISC reports (0 items):** The current code queries only the `riscreports` division a... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-RESEARCH.md | Markdown | 27ea3c2bdcee89523de64fb39e87e0026746d5f1d1bebe49352ae780a4282fc0 | 1 | 896 |
needed.
---
## Architecture Patterns
### Recommended Project Structure
New files this phase adds:
```
src/jku_kb/scrapers/
└── openalex_xref.py # Shared cross-reference utility (NEW)
tests/unit/test_scrapers/
└── test_openalex_xref.py # Tests for the shared utility (NEW)
tests/fixtures/
└── opencast_pa... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-RESEARCH.md | Markdown | 277f427e06cce8909f89dcd719aa4b39dab31ab5979def83d35f4ba168051dd4 | 2 | 896 |
`RawItem.metadata`, which is NOT persisted to `manifest.jsonl`.
**The actual source of external ID data is the OpenAlex scraper's discover output.** Two options for the utility:
**Option A (recommended):** Have the OpenAlex scraper write a secondary `external_ids.jsonl` file during discover, containing `{item_id, arx... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-RESEARCH.md | Markdown | 7c9dba5e30d78df8ffe1fcb579ecab2767d0edabbb50504accb57a8005bfbfd6 | 3 | 896 |
`.find_parent("tr")` + `.find_all("td")` | Already imported in pdf_fetch.py |
| Rate limiting | Manual sleep loops | `BaseScraper._rate_limited_get()` | Already handles arXiv 0.33/s and S2 0.33/s via `DOMAIN_RATE_LIMITS` |
| URL normalisation | Custom urljoin | `urllib.parse.urljoin` | Already used in pdf_fetch.py |
| ... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-RESEARCH.md | Markdown | 77d707e86451223c1364bb4254b519f396936dbb210bbbee0f163e2793011e10 | 4 | 896 |
consistently. Already handled for `result` list (the `_extract_results` function handles dict/list/missing).
**Warning signs:** TypeError on `total > offset` comparison.
### Pitfall 5: RISC listing page link deduplication
**What goes wrong:** The same PDF appears on both `riscreports` and `techreports` division pages.... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-RESEARCH.md | Markdown | 3d1bce467602f3a56e3e494738a1414683d0d36d48ed16067b268eba374111a4 | 5 | 896 |
"title,authors,year,abstract,externalIds,openAccessPdf,url"},
)
if response.status_code == 200:
return _paper_to_raw_item(response.json())
return None
```
---
## State of the Art
| Old Approach | Current Approach | When Changed | Impact |
|--------------|------------------|--------------|--------... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-RESEARCH.md | Markdown | 95371d29206383a0aa297ec1c118b111d7e701efa174632f8f316ad9b2cf01b4 | 6 | 896 |
, "doi": "10.5678/other"}
```
**Pipeline ordering:** The orchestrator's `run_phase_discover` accepts `source_ids` list. When running all sources, OpenAlex must run before arXiv and Semantic Scholar. This is handled by running discover in two passes:
1. Pass 1: `["openalex_jku"]` — generates `discover_xref.jsonl`
2. Pa... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-RESEARCH.md | Markdown | ad02d764af4962489c1bc90b4b2a49faabc8b0c383c4a530a5684f929e47aa12 | 7 | 896 |
and just need additional test methods.)*
---
## Sources
### Primary (HIGH confidence)
- Direct code reading: `src/jku_kb/scrapers/pdf_fetch.py`, `opencast.py`, `arxiv.py`, `semantic_scholar.py`, `openalex.py`, `base.py`
- Direct code reading: `tests/unit/test_scrapers/test_pdf_fetch.py`, `test_opencast.py`, `test_ar... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-RESEARCH.md | Markdown | 7969136495ee06a13c350780912cb35784251305ffcd12ffcfbdc5d683db9df1 | 8 | 363 |
---
phase: 14
slug: api-scraper-fixes
status: draft
nyquist_compliant: false
wave_0_complete: false
created: 2026-03-23
---
# Phase 14 — Validation Strategy
> Per-phase validation contract for feedback sampling during execution.
---
## Test Infrastructure
| Property | Value |
|----------|-------|
| **Framework** |... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-VALIDATION.md | Markdown | e71cb7f365b044672c8193a0defc031e2cea17eff27477f077d37272c2bc8c04 | 0 | 896 |
then manually verify with `curl "https://media.jku.at/search/episode.json?limit=10&offset=0"` |
| RISC report count >= 50 | SRC-01 | Requires live HTTP access to RISC website | Run discover against live site, check manifest count |
| arXiv result count >= 500 | SRC-03 | Requires live arXiv API access | Run discover, ve... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-VALIDATION.md | Markdown | 95645b1499b8ec2e8d6183e45388703f214a69cbaa576b17e953569f1129b923 | 1 | 196 |
---
phase: 14-api-scraper-fixes
verified: 2026-03-23T20:15:00Z
status: passed
score: 16/16 must-haves verified
re_verification: false
---
# Phase 14: API Scraper Fixes Verification Report
**Phase Goal:** The four broken or massively underperforming API scrapers each yield item counts at or above their expected minimu... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-VERIFICATION.md | Markdown | d77a7309912f3997c626bcd66ae38af299abe9fceed1a6222d777f6a24378b0f | 0 | 896 |
cache_dir)` then `_discover_by_dois()` at lines 33-38 of semantic_scholar.py |
| 15 | Semantic Scholar scraper uses /paper/batch endpoint for efficient cross-ref lookups | VERIFIED | `S2_BATCH_BASE = "https://api.semanticscholar.org/graph/v1/paper/batch"` at line 14; `_rate_limited_post(S2_BATCH_BASE, ...)` in `_discov... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-VERIFICATION.md | Markdown | 6c15ba9b3c77ab76f2dd2643ff3a1ef8f47e0073eba9fc074a56d62e3bdf4ea2 | 1 | 896 |
`openalex_xref.py` | `discover_xref.jsonl` | orjson.loads line-by-line reading | WIRED | `entry = orjson.loads(line)` in `read_openalex_xref()` at line 37 |
| `arxiv.py` | `openalex_xref.py` | read_openalex_xref import | WIRED | `from jku_kb.scrapers.openalex_xref import _normalize_arxiv_id, read_openalex_xref` at line... | jku-encyclopedia | .planning/phases/14-api-scraper-fixes/14-VERIFICATION.md | Markdown | 7b681a382ccba9c344d4136f299d3953a6690cb4d7ae2b7e57bd9aa6a73eb002 | 2 | 849 |
---
phase: 15-verification-completeness-reporting
plan: 01
type: execute
wave: 1
depends_on: []
files_modified:
- data/sources.json
- src/jku_kb/cli.py
- tests/unit/test_cli.py
autonomous: true
requirements:
- REPORT-01
- REPORT-02
- REPORT-03
- REPORT-04
must_haves:
truths:
- "Running `jku-kb veri... | jku-encyclopedia | .planning/phases/15-verification-completeness-reporting/15-01-PLAN.md | Markdown | b0aaaa9ccee945035c81e525b81ea90ebf716ded7a1cdf43d836e87b1607aa51 | 0 | 896 |
SourceEntry]) -> SourceQuality: ...
async def run_health_probes(entries: list[SourceEntry], source_ids: list[str] | None = None, overall_timeout: float = 30.0) -> list[HealthResult]: ...
```
From src/jku_kb/scrapers/__init__.py:
```python
_REGISTRY: dict[str, type] = {}
def get_scraper(source_id: str, settings: Settin... | jku-encyclopedia | .planning/phases/15-verification-completeness-reporting/15-01-PLAN.md | Markdown | 669029ed91beacee0f958aba25a70c3f7d33b7037ee5584d2ca0694ed550bd7f | 1 | 896 |
to 1 as a placeholder. Plan 02 will update these with live-derived values (~80% of actual count).
2. **Add quality blocks** with `expected_min_items: 0` to all 34 entries that currently lack a quality block. These are informational/placeholder entries without scrapers. For each entry missing a `"quality"` key, add:
... | jku-encyclopedia | .planning/phases/15-verification-completeness-reporting/15-01-PLAN.md | Markdown | cb0b5fce424867bd2334c14ebfbcfd79c6e203fbc6c15244ef97c5f5d11df5e5 | 2 | 896 |
": "[yellow]warn[/yellow]",
"fail": "[red]FAIL[/red]",
"error": "[red]ERROR[/red]",
"pending": "[dim]...[/dim]",
}
def _build_verify_table(
rows: list[dict[str, Any]],
*,
title: str = "Source Verification",
) -> Table:
"""Build Rich table for verify command output."""
table = Table(
... | jku-encyclopedia | .planning/phases/15-verification-completeness-reporting/15-01-PLAN.md | Markdown | 3425de797cede801565d04c88f2bba3d3718b3d9aa54eb34f1f2f53cb166310e | 3 | 896 |
except Exception as exc:
return {"source_id": sid, "actual": 0, "error": str(exc)}
# Use Rich Live for real-time progress (non-JSON mode)
pending_rows: dict[str, dict[str, Any]] = {
sid: {"source_id": sid, "status": "pending", "expected": 0, "actual": 0, "error": None}
... | jku-encyclopedia | .planning/phases/15-verification-completeness-reporting/15-01-PLAN.md | Markdown | d46bb7ada4beca5f47180daba5620926199efbca739fc52d5deadfd8eb8a83ce | 4 | 896 |
, the source is marked FAIL with error captured, and verification continues for remaining sources.
- The `--source` flag accepts a single source ID (not "all") for quick development testing.
</action>
<verify>
<automated>python -c "from jku_kb.cli import app; cmds=[c.name for c in app.registered_commands]; asse... | jku-encyclopedia | .planning/phases/15-verification-completeness-reporting/15-01-PLAN.md | Markdown | 43f35de6a9079ec6de044d676cbc37b6a3871f95dcde0becf8eea566b53805f8 | 5 | 896 |
mock_entries),
patch("jku_kb.cli.run_health_probes", new_callable=AsyncMock, return_value=mock_probes),
patch("jku_kb.cli._get_discoverable_source_ids", return_value=["test_src"]),
patch("jku_kb.cli.get_quality_for_source", return_value=SourceQuality(expected_min_items=10)),
patch("jku_k... | jku-encyclopedia | .planning/phases/15-verification-completeness-reporting/15-01-PLAN.md | Markdown | bf37edb6382588222fa5a0378ce37d0a5be43ed134b4d03b4f5f2c91fd436bb8 | 6 | 896 |
patch
from typer.testing import CliRunner
from jku_kb.cli import app
from jku_kb.models import HealthResult, SourceEntry, SourceQuality
runner = CliRunner()
mock_entries = [
SourceEntry(
source_id="single_src",
name="Single Test",
category="test",
... | jku-encyclopedia | .planning/phases/15-verification-completeness-reporting/15-01-PLAN.md | Markdown | 4e58c75388e4090bed75a6f20c540f0afb89d368080cad8cd52257f1828b1414 | 7 | 896 |
(`
- tests/unit/test_cli.py does NOT contain `def test_health_command_table(`
- tests/unit/test_cli.py does NOT contain `def test_health_command_no_manifest(`
- tests/unit/test_cli.py contains `def test_compute_health_status_` (kept, not removed)
- `python -m pytest tests/unit/test_cli.py -x -q` exits w... | jku-encyclopedia | .planning/phases/15-verification-completeness-reporting/15-01-PLAN.md | Markdown | 579de73e6ca152bbfda98ecf6b3bde2e78816d139ee5c945e5a459292c322138 | 8 | 404 |
---
phase: 15-verification-completeness-reporting
plan: 01
subsystem: cli
tags: [typer, rich, asyncio, verify, sources.json, quality-gates]
# Dependency graph
requires:
- phase: 14-api-scraper-completion
provides: All API scrapers registered in _REGISTRY with discover() method
- phase: 13-institute-web-crawler... | jku-encyclopedia | .planning/phases/15-verification-completeness-reporting/15-01-SUMMARY.md | Markdown | 42f3bf23392df0791aee181f8a53bc5a13cfa445efa76176e5a431304c5b3452 | 0 | 896 |
only)
## Deviations from Plan
### Auto-fixed Issues
**1. [Rule 1 - Bug] Fixed JSON output corruption from Rich console.print**
- **Found during:** Task 3 (test_verify_command_discover_error)
- **Issue:** `console.print(orjson.dumps(...).decode())` injected ANSI escape codes into JSON output, causing `json.JSONDecode... | jku-encyclopedia | .planning/phases/15-verification-completeness-reporting/15-01-SUMMARY.md | Markdown | 53f807d0f7263b05b4a735b38cd5e88bea3629b5cf437343e1f8e94a4007441e | 1 | 300 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.