feat(excel2xml): support 2-digit-years in find_date_in_string() (DEV-…

…2633) (#711)

src/dsp_tools/commands/excel2xml/excel2xml_lib.py

@@ -94,19 +94,24 @@ def _find_french_bc_date(
 
 def find_date_in_string(string: str) -> Optional[str]:
     """
-    Checks if a string contains a date value (single date, or date range), and returns the first found date as
-    DSP-formatted string. Returns None if no date was found.
+    Checks if a string contains a date value (single date, or date range),
+    and returns the first found date as DSP-formatted string.
+    Returns None if no date was found.
 
     Notes:
         - All dates are interpreted in the Christian era and the Gregorian calendar.
         - BC dates are only supported in French notation (e.g. 1000-900 av. J.-C.).
         - The years 0000-2999 are supported, in 3/4-digit form.
         - Dates written with slashes are always interpreted in a European manner: 5/11/2021 is the 5th of November.
+        - In the European notation, 2-digit years are expanded to 4 digits, with the current year as watershed:
+            - 30.4.24 -> 30.04.2024
+            - 30.4.25 -> 30.04.1925
 
     Currently supported date formats:
         - 0476-09-04 -> GREGORIAN:CE:0476-09-04:CE:0476-09-04
         - 0476_09_04 -> GREGORIAN:CE:0476-09-04:CE:0476-09-04
         - 30.4.2021 -> GREGORIAN:CE:2021-04-30:CE:2021-04-30
+        - 30.4.21 -> GREGORIAN:CE:2021-04-30:CE:2021-04-30
         - 5/11/2021 -> GREGORIAN:CE:2021-11-05:CE:2021-11-05
         - Jan 26, 1993 -> GREGORIAN:CE:1993-01-26:CE:1993-01-26
         - 28.2.-1.12.1515 -> GREGORIAN:CE:1515-02-28:CE:1515-12-01
@@ -175,6 +180,7 @@ def _find_date_in_string_throwing(string: str) -> str | None:
     This function is the same as find_date_in_string(), but may raise a ValueError instead of returning None.
     """
     year_regex = r"([0-2]?[0-9][0-9][0-9])"
+    year_regex_2_or_4_digits = r"((?:[0-2]?[0-9])?[0-9][0-9])"
     month_regex = r"([0-1]?[0-9])"
     day_regex = r"([0-3]?[0-9])"
     sep_regex = r"[\./]"
@@ -186,25 +192,31 @@ def _find_date_in_string_throwing(string: str) -> str | None:
 
     # template: 2021-01-01 | 2015_01_02
     iso_date = regex.search(rf"{lookbehind}{year_regex}[_-]([0-1][0-9])[_-]([0-3][0-9]){lookahead}", string)
+
     # template: 6.-8.3.1948 | 6/2/1947 - 24.03.1948
     eur_date_range_regex = (
         rf"{lookbehind}"
-        rf"{day_regex}{sep_regex}(?:{month_regex}{sep_regex}{year_regex}?)? ?(?:-|:|to) ?"
-        rf"{day_regex}{sep_regex}{month_regex}{sep_regex}{year_regex}"
+        rf"{day_regex}{sep_regex}(?:{month_regex}{sep_regex}{year_regex_2_or_4_digits}?)? ?(?:-|:|to) ?"
+        rf"{day_regex}{sep_regex}{month_regex}{sep_regex}{year_regex_2_or_4_digits}"
         rf"{lookahead}"
     )
     eur_date_range = regex.search(eur_date_range_regex, string)
+
     # template: 1.4.2021 | 5/11/2021
+    eur_date_regex = rf"{lookbehind}{day_regex}{sep_regex}{month_regex}{sep_regex}{year_regex_2_or_4_digits}{lookahead}"
     eur_date = regex.search(
-        rf"{lookbehind}{day_regex}{sep_regex}{month_regex}{sep_regex}{year_regex}{lookahead}",
+        eur_date_regex,
         string,
     )
+
     # template: March 9, 1908 | March5,1908 | May 11, 1906
     all_months = "|".join(_months_dict)
     monthname_date_regex = rf"{lookbehind}({all_months}) ?{day_regex}, ?{year_regex}{lookahead}"
     monthname_date = regex.search(monthname_date_regex, string)
+
     # template: 1849/50 | 1849-50 | 1849/1850
     year_range = regex.search(lookbehind + year_regex + r"[/-](\d{1,4})" + lookahead, string)
+
     # template: 1907
     year_only = regex.search(rf"{lookbehind}{year_regex}{lookahead}", string)
 
@@ -233,13 +245,25 @@ def _from_iso_date(iso_date: Match[str]) -> str:
     return f"GREGORIAN:CE:{date.isoformat()}:CE:{date.isoformat()}"
 
 
+def _expand_2_digit_year(year: int) -> int:
+    current_year = datetime.date.today().year - 2000
+    if year <= current_year:
+        return year + 2000
+    elif year <= 99:
+        return year + 1900
+    else:
+        return year
+
+
 def _from_eur_date_range(eur_date_range: Match[str]) -> str:
     startday = int(eur_date_range.group(1))
     startmonth = int(eur_date_range.group(2)) if eur_date_range.group(2) else int(eur_date_range.group(5))
     startyear = int(eur_date_range.group(3)) if eur_date_range.group(3) else int(eur_date_range.group(6))
+    startyear = _expand_2_digit_year(startyear)
     endday = int(eur_date_range.group(4))
     endmonth = int(eur_date_range.group(5))
     endyear = int(eur_date_range.group(6))
+    endyear = _expand_2_digit_year(endyear)
     startdate = datetime.date(startyear, startmonth, startday)
     enddate = datetime.date(endyear, endmonth, endday)
     if enddate <= startdate:
@@ -251,6 +275,7 @@ def _from_eur_date(eur_date: Match[str]) -> str:
     startday = int(eur_date.group(1))
     startmonth = int(eur_date.group(2))
     startyear = int(eur_date.group(3))
+    startyear = _expand_2_digit_year(startyear)
     date = datetime.date(startyear, startmonth, startday)
     return f"GREGORIAN:CE:{date.isoformat()}:CE:{date.isoformat()}"
 

test/unittests/commands/excel2xml/test_excel2xml_lib.py

@@ -1,3 +1,4 @@
+import datetime
 import unittest
 import warnings
 from pathlib import Path
@@ -204,13 +205,26 @@ def test_find_date_in_string_eur_date(self) -> None:
         assert excel2xml.find_date_in_string("x -2193_01_26- x") == "GREGORIAN:CE:2193-01-26:CE:2193-01-26"
         assert excel2xml.find_date_in_string("x 2193_02_30 x") is None
 
+    def test_find_date_in_string_eur_date_2_digit(self) -> None:
+        cur = str(datetime.date.today().year - 2000)  # in 2024, this will be "24"
+        nxt = str(datetime.date.today().year - 2000 + 1)  # in 2024, this will be "25"
+        assert excel2xml.find_date_in_string(f"x 30.4.{cur} x") == f"GREGORIAN:CE:20{cur}-04-30:CE:20{cur}-04-30"
+        assert excel2xml.find_date_in_string(f"x 30.4.{nxt} x") == f"GREGORIAN:CE:19{nxt}-04-30:CE:19{nxt}-04-30"
+        assert excel2xml.find_date_in_string(f"x 31.4.{nxt} x") is None
+
     def test_find_date_in_string_eur_date_range(self) -> None:
         """template: 27.-28.1.1900"""
         assert excel2xml.find_date_in_string("x 25.-26.2.0800 x") == "GREGORIAN:CE:0800-02-25:CE:0800-02-26"
         assert excel2xml.find_date_in_string("x 25. - 26.2.0800 x") == "GREGORIAN:CE:0800-02-25:CE:0800-02-26"
         assert excel2xml.find_date_in_string("x 25.-24.2.0800 x") is None
         assert excel2xml.find_date_in_string("x 25.-25.2.0800 x") is None
 
+    def test_find_date_in_string_eur_date_range_2_digit(self) -> None:
+        cur = str(datetime.date.today().year - 2000)  # in 2024, this will be "24"
+        nxt = str(datetime.date.today().year - 2000 + 1)  # in 2024, this will be "25"
+        assert excel2xml.find_date_in_string(f"x 15.-16.4.{cur} x") == f"GREGORIAN:CE:20{cur}-04-15:CE:20{cur}-04-16"
+        assert excel2xml.find_date_in_string(f"x 15.-16.4.{nxt} x") == f"GREGORIAN:CE:19{nxt}-04-15:CE:19{nxt}-04-16"
+
     def test_find_date_in_string_eur_date_range_across_month(self) -> None:
         """template: 26.2.-24.3.1948"""
         assert excel2xml.find_date_in_string("x _1.3. - 25.4.2022_ x") == "GREGORIAN:CE:2022-03-01:CE:2022-04-25"
@@ -220,6 +234,12 @@ def test_find_date_in_string_eur_date_range_across_month(self) -> None:
         assert excel2xml.find_date_in_string("x 28.2.-26.2.1515 x") is None
         assert excel2xml.find_date_in_string("x 28.2.-28.2.1515 x") is None
 
+    def test_find_date_in_string_eur_date_range_across_month_2_digit(self) -> None:
+        cur = str(datetime.date.today().year - 2000)  # in 2024, this will be "24"
+        nxt = str(datetime.date.today().year - 2000 + 1)  # in 2024, this will be "25"
+        assert excel2xml.find_date_in_string(f"x 15.04.-1.5.{cur} x") == f"GREGORIAN:CE:20{cur}-04-15:CE:20{cur}-05-01"
+        assert excel2xml.find_date_in_string(f"x 15.04.-1.5.{nxt} x") == f"GREGORIAN:CE:19{nxt}-04-15:CE:19{nxt}-05-01"
+
     def test_find_date_in_string_eur_date_range_across_year(self) -> None:
         """template: 1.12.1973 - 6.1.1974"""
         assert excel2xml.find_date_in_string("x 1.9.2022-3.1.2024 x") == "GREGORIAN:CE:2022-09-01:CE:2024-01-03"
@@ -231,6 +251,12 @@ def test_find_date_in_string_eur_date_range_across_year(self) -> None:
         assert excel2xml.find_date_in_string("x 25/12/2022-03/01/2022 x") is None
         assert excel2xml.find_date_in_string("x 25/12/2022-25/12/2022 x") is None
 
+    def test_find_date_in_string_eur_date_range_across_year_2_digit(self) -> None:
+        cur = str(datetime.date.today().year - 2000)  # in 2024, this will be "24"
+        nxt = str(datetime.date.today().year - 2000 + 1)  # in 2024, this will be "25"
+        assert excel2xml.find_date_in_string(f"x 15.04.23-1.5.{cur} x") == f"GREGORIAN:CE:2023-04-15:CE:20{cur}-05-01"
+        assert excel2xml.find_date_in_string(f"x 15.04.{nxt}-1.5.26 x") == f"GREGORIAN:CE:19{nxt}-04-15:CE:1926-05-01"
+
     def test_find_date_in_string_monthname(self) -> None:
         """template: February 9, 1908 | Dec 5,1908"""
         assert excel2xml.find_date_in_string("x Jan 26, 1993 x") == "GREGORIAN:CE:1993-01-26:CE:1993-01-26"