Merge pull request #173 from mlibrary/citation-trailing-slashes

citations and compress api output

Author: niquerio

api/catalog_api/main.py

@@ -1,4 +1,5 @@
 from fastapi import FastAPI, HTTPException
+from fastapi.middleware.gzip import GZipMiddleware
 from prometheus_fastapi_instrumentator import Instrumentator
 from prometheus_client import Histogram
 
@@ -9,6 +10,7 @@
 app = FastAPI(
     title="Catalog Search API", description="REST API for Catalog Search Solr"
 )
+app.add_middleware(GZipMiddleware, minimum_size=1000)
 
 Instrumentator().instrument(app).expose(app)
 

api/catalog_api/marc.py

@@ -5,6 +5,10 @@
 from collections.abc import Callable
 from catalog_api.entities import SearchField, FieldElement, PairedField
 
+# used with rstrip_chars. When we want to trim the end of a string, these are
+# the ones that usually should be removed.
+TRIM_CHARS = "/.,: "
+
 
 class Linkage:
     def __init__(self, field: pymarc.Field):
@@ -32,13 +36,16 @@ class FieldRuleset:
     search: list | None = None
     browse_sfs: str | None = None
     filter: Callable[..., bool] = lambda field: True
+    rstrip_chars: str = ""
 
     def has_any_subfields(self, field: pymarc.Field) -> bool:
         return bool(self._get_subfields(field, self.text_sfs))
 
     def value_for(self, field: pymarc.Field):
         result = {
-            "text": self._get_subfields(field, self.text_sfs).strip(),
+            "text": self._get_subfields(field, self.text_sfs)
+            .rstrip(self.rstrip_chars)
+            .strip(),
             "tag": field.tag,
         }
 

api/catalog_api/record.py

@@ -1,7 +1,7 @@
 from __future__ import annotations
 from catalog_api.solr_client import SolrClient
 from catalog_api.solr import SolrDocProcessor
-from catalog_api.marc import Processor, FieldRuleset
+from catalog_api.marc import Processor, FieldRuleset, TRIM_CHARS
 import re
 import pymarc
 import io
@@ -585,6 +585,7 @@ class TaggedCitation:
             "ruleset": FieldRuleset(
                 tags=["100", "101", "110", "111", "700", "710", "711"],
                 text_sfs="abcdefgjklnpqtu4",
+                rstrip_chars=TRIM_CHARS,
             ),
             "ris": ["AU"],
             "meta": ["author"],
@@ -613,6 +614,7 @@ class TaggedCitation:
                 filter=lambda field: (
                     field.indicator1 == "0" and re.match("ed", field.get("e", ""))
                 ),
+                rstrip_chars=TRIM_CHARS,
             ),
             "ris": ["ED", "A2"],
             "meta": ["editor"],
@@ -626,8 +628,7 @@ class TaggedCitation:
         {
             "kind": "marc",
             "ruleset": FieldRuleset(
-                tags=["245"],
-                text_sfs="abnp",
+                tags=["245"], text_sfs="abnp", rstrip_chars=TRIM_CHARS
             ),
             "ris": ["JF", "T1", "TI"],  # JF seems sus; how do we know it's a journal?
             "meta": ["title", "journal_title"],
@@ -697,6 +698,7 @@ class TaggedCitation:
             "ruleset": FieldRuleset(
                 tags=["264", "260"],
                 text_sfs="b",
+                rstrip_chars=TRIM_CHARS,
             ),
             "ris": ["PB"],
             "meta": ["publisher"],
@@ -1078,10 +1080,7 @@ def type(self):
     @property
     def title(self):
         rulesets = (
-            FieldRuleset(
-                tags=["245"],
-                text_sfs="abp",
-            ),
+            FieldRuleset(tags=["245"], text_sfs="abp", rstrip_chars=TRIM_CHARS),
         )
         return self._get_marc_content(rulesets)
 
@@ -1113,11 +1112,13 @@ def publisher_place(self):
             FieldRuleset(
                 tags=["260"],
                 text_sfs="a",
+                rstrip_chars=TRIM_CHARS,
             ),
             FieldRuleset(
                 tags=["264"],
                 text_sfs="a",
                 filter=lambda field: (field.indicator2 == "1"),
+                rstrip_chars=TRIM_CHARS,
             ),
         )
         return self._get_marc_content(rulesets)
@@ -1128,11 +1129,13 @@ def publisher(self):
             FieldRuleset(
                 tags=["260"],
                 text_sfs="b",
+                rstrip_chars=TRIM_CHARS,
             ),
             FieldRuleset(
                 tags=["264"],
                 text_sfs="b",
                 filter=lambda field: (field.indicator2 == "1"),
+                rstrip_chars=TRIM_CHARS,
             ),
         )
         return self._get_marc_content(rulesets)
@@ -1163,13 +1166,15 @@ def author(self):
                 filter=lambda field: (
                     field.indicator1 == "1" and field.get("e") not in field_e_strings
                 ),
+                rstrip_chars=TRIM_CHARS,
             ),
             FieldRuleset(
                 tags=["100", "700"],
                 text_sfs="ab",
                 filter=lambda field: (
                     field.indicator1 == "0" and field.get("e") not in field_e_strings
                 ),
+                rstrip_chars=TRIM_CHARS,
             ),
         )
 
@@ -1180,6 +1185,7 @@ def author(self):
                 tags=["110", "111", "710", "711"],
                 text_sfs="ab",
                 filter=lambda field: (field.get("e") not in field_e_strings),
+                rstrip_chars=TRIM_CHARS,
             ),
         )
         corporate_authors = self._to_literal(
@@ -1201,13 +1207,15 @@ def editor(self):
                 filter=lambda field: (
                     field.indicator1 == "1" and field.get("e") in field_e_strings
                 ),
+                rstrip_chars=TRIM_CHARS,
             ),
             FieldRuleset(
                 tags=["700"],
                 text_sfs="ab",
                 filter=lambda field: (
                     field.indicator1 == "0" and field.get("e") in field_e_strings
                 ),
+                rstrip_chars=TRIM_CHARS,
             ),
         )
         result = self._to_author(self._get_marc_contents(rulesets))

api/tests/test_record.py

@@ -1519,6 +1519,12 @@ def test_title(self):
 
         assert subject.title == "a b p"
 
+    def test_title_does_not_have_trailing_chars(self):
+        record = create_record_with_paired_field(tag="245")
+        record["245"]["p"] = "p . /,: / "
+        subject = CSL(marc_record=record)
+        assert subject.title == "a b p"
+
     def test_call_number(self, solr_bib):
         solr_bib["callnumber"].append("some other call number")
         subject = CSL(solr_doc=solr_bib)