deploy: b4420d4

Author: tlunet

bin/README.md

@@ -0,0 +1,19 @@
+# Script development guidelines
+
+## Testing your changes
+
+Before pushing you changes, create a `test.sh` file in this folder with this content :
+
+```sh
+#!/bin/bash
+
+BODY="" # -> put there the content of the issue
+
+python $SCRIPT_NAME -b "${BODY}"
+```
+
+- from github, copy-paste the issue body using the "..." icon and `Copy Markdown`
+- replace $SCRIPT_NAME by the script you want to test, like `issue_to_bibtex.py`, `arxiv_to_publications_correct.py`, ...
+
+Then, check that the modification of the `_bibliography/pint.bib` file is correct.
+You can commit the changes on the `pint.bib` file (done now, no need to do it again), or reset those and trigger the scripts on Github using the label manipulation on the issue (check if the whole pipeline works well).

bin/arxiv_to_publications_correct.py

@@ -8,26 +8,7 @@
 from bibtexparser.bwriter import BibTexWriter
 from requests.exceptions import RequestException
 
-
-def normalize_month_fields(bib):
-    def replace_month(match):
-        prefix, value = match.groups()
-        value = unescape(value).strip()
-        if value.startswith("{") and value.endswith("}"):
-            return match.group(0)
-        if value.startswith('"') and value.endswith('"'):
-            value = value[1:-1].strip()
-        value = value.strip("'\"{}")
-        value = re.sub(r"[^A-Za-z]", "", value)
-        return f"{prefix}{{{value}}},"
-
-    return re.sub(
-        r"(^\s*month\s*=\s*)([^,\n]+)\s*,",
-        replace_month,
-        bib,
-        flags=re.MULTILINE | re.IGNORECASE,
-    )
-
+from utils import fixBadBibFormat
 
 def fetch_doi_content(url, accept_header, description):
     try:
@@ -104,7 +85,7 @@ def fetch_doi_content(url, accept_header, description):
             if id != id_db:
                 print(f'Note: ID updated from {id_db} to {id} to reflect the publication year.')
             bib = "{".join([bType] + [','.join([id]+rest2)] + rest1[1:])
-            bib = normalize_month_fields(bib)
+            bib = fixBadBibFormat(bib)
             bib_db = bibtexparser.loads(bib)
             new_entries = bib_db.get_entry_list()
             if not new_entries:

bin/issue_to_bibtex.py

@@ -11,19 +11,7 @@
 from bibtexparser.bibdatabase import BibDatabase
 from bibtexparser.bwriter import BibTexWriter
 
-
-def fixBadBibFormat(bibString):
-    bibString = bibString.strip()
-    assert bibString.startswith("@") and bibString.endswith("}"), \
-        "bib entry should start with '@' and finish with '}', got :\n{bibString}"
-    content = bibString[1:-1]
-    fields = content.split(",")
-    for i, field in enumerate(fields[1:]):
-        item = field.split("=")
-        if len(item) == 2 and "{" not in item[1]:
-            item[1] = "{"+item[1]+"}"
-            fields[i+1] = "=".join(item)
-    return "@"+",".join(fields)+"}"
+from utils import fixBadBibFormat
 
 try:
 

bin/request.txt

@@ -0,0 +1,30 @@
+def fixBadBibFormat(bibString:str)->str:
+    """
+    Fixes the formatting of a BibTeX entry string by ensuring all field values are enclosed in braces.
+
+    Parameters
+    ----------
+    bibString : str
+        The BibTeX entry string to be fixed. Must start with '@' and end with '}'.
+
+    Returns
+    -------
+    str
+        The corrected BibTeX entry string with all field values enclosed in braces.
+
+    Raises
+    ------
+    AssertionError
+        If the input string does not start with '@' or does not end with '}'.
+    """
+    bibString = bibString.strip()
+    assert bibString.startswith("@") and bibString.endswith("}"), \
+        "bib entry should start with '@' and finish with '}', got :\n{bibString}"
+    content = bibString[1:-1]
+    fields = content.split(",")
+    for i, field in enumerate(fields[1:]):
+        item = field.split("=")
+        if len(item) == 2 and "{" not in item[1]:
+            item[1] = "{"+item[1]+"}"
+            fields[i+1] = "=".join(item)
+    return "@"+",".join(fields)+"}"