diff options
Diffstat (limited to 'tools/extract_missalemeum_oracle.py')
| -rw-r--r-- | tools/extract_missalemeum_oracle.py | 58 |
1 files changed, 57 insertions, 1 deletions
diff --git a/tools/extract_missalemeum_oracle.py b/tools/extract_missalemeum_oracle.py index a010b52..4b010e2 100644 --- a/tools/extract_missalemeum_oracle.py +++ b/tools/extract_missalemeum_oracle.py @@ -18,7 +18,7 @@ # exists for every day. # # One line per day, pipe-separated: -# date|rank|colors|title|tempora|commemorations|displaced|n_masses|commemoration_ids +# date|rank|colors|title|tempora|commemorations|displaced|n_masses|commemoration_ids|first|gospel # # - rank/colors/title/tempora/commemorations/displaced are info.rank, # info.colors (sorted, concatenated, e.g. "pv"), info.title, info.tempora @@ -42,6 +42,35 @@ # is ever commemorated there). Never parsed for its date component by this # fixture or the comparator (info.rank/colors already give the day's own # values); kept opaque and compared as a plain string. +# - first/gospel (Task 9) are the day's own Epistle/Lesson and Gospel +# citations, extracted from entry[0]["sections"] -- the section with +# id "Lectio" ("Epistle") and "Evangelium" ("Gospel") respectively. Each +# section's own body is a [[english, latin]] pair; the ENGLISH text +# (body[0][0]) embeds the citation as one of possibly several +# "*...*"-wrapped (markdown-italic) spans -- e.g. "Lesson from the letter +# of St. Paul... \n*Titus 2:11-15*\nBeloved: ...". It is NOT always the +# FIRST such span: Holy Saturday's own Gospel section opens with a rubric +# note, itself "*"-wrapped ("*While singing the Gospel candles are not +# being hold.*"), before the real citation "*Matt 28:1-7*" -- checked +# directly against the raw JSON, not assumed. The citation span is +# identified structurally instead of positionally: the first "*...*" span +# under 40 characters that contains a chapter:verse-shaped digit pair +# (`\d+\s*[:,.]\s*\d+`, matching all three separators actually used in +# this source -- "Titus 2:11-15", "4 Kings, 5:1-15", "John 20. 19-31"). +# Verified exhaustively over the whole fixture (all 1 458 Lectio/ +# Evangelium sections across the 728 days that carry one): exactly one +# candidate per section, zero ambiguous, zero false positives from a +# longer rubric sentence. TWO days (Good Friday, both years -- the "Missa +# Praesanctificatorum" liturgy) have no "Lectio"/"Evangelium" section at +# all (a multi-lesson structure instead, "Lectiones"/"Passio", with no +# single reading occupying the Epistle/Gospel slot this schema assumes, +# the SAME shape colitur's own test_lectionary.ml records for its own +# hand-authored Holy Week data) -- "-" for both fields on those two rows, +# a WARNING on stderr naming the date, never a silent guess. Extracted +# VERBATIM (only .strip()ped of surrounding whitespace) -- not normalized, +# not re-punctuated: the comparator's job, not this extractor's, matching +# the same "verbatim, not translated or slugified" discipline the fields +# above already state. # - Spaces in tempora are turned to "_" (matching the "no field has an # internal space" convention test/fixtures/lectio-ef-2005-2050.txt already # uses, so this fixture can be read the same simple way -- split on '|', @@ -50,8 +79,27 @@ # pass-through display, is space-collapsed for a cheap column-count check). import json import os +import re import sys +STAR_RE = re.compile(r"\*([^*]+)\*") +VERSE_RE = re.compile(r"\d+\s*[:,.]\s*\d+") + + +def extract_citation(sections, section_id, date, label): + for sec in sections: + if sec["id"] != section_id: + continue + text = sec["body"][0][0] + candidates = [s.strip() for s in STAR_RE.findall(text) if VERSE_RE.search(s) and len(s) < 40] + if candidates: + return candidates[0] + print(f"WARNING: {date}: {section_id!r} section found but no citation-shaped span in it ({label})", + file=sys.stderr) + return "-" + print(f"WARNING: {date}: no {section_id!r} section ({label})", file=sys.stderr) + return "-" + def main(): if len(sys.argv) != 3: @@ -84,8 +132,16 @@ def main(): comms = ";".join(comm_titles) or "-" disp = ";".join(disp_titles) or "-" ids = ";".join(comm_ids) or "-" + sections = data[0]["sections"] + cit_first = extract_citation(sections, "Lectio", date, "Epistle") + cit_gospel = extract_citation(sections, "Evangelium", date, "Gospel") + for t in (cit_first, cit_gospel): + if "|" in t: + print(f"ERROR: {date}: citation {t!r} contains a delimiter this fixture uses", file=sys.stderr) + return 1 rows.append( f"{date}|{first['rank']}|{colors}|{first['title']}|{tempora}|{comms}|{disp}|{len(data)}|{ids}" + f"|{cit_first}|{cit_gospel}" ) if len(rows) != 730: |