gh-155292: Skip updating unicodedata with mismatched interpreter

mkstringprep uses things like str.lower(), so it generates the wrong
result if run in an interpreter with a different Unicode data version
than the target.

This means that updating the Unicode version is a two-step process:
run makeunicodedata.py, then compile, then run mkstringprep.py.

The two steps can (and should) be combined when re-running
regen-unicodedata to verify that the data is up to date.
The GH-155292 fix only considered that case.

Change makeunicodedata.py to only run mkstringprep.py when the
current interpreter is up to it. Otherwise, show a reminder.

As an extra complication, download the input (RFC 3454) in
the "first step", since an out-of-date stringprep.py's freshness
assertion may prevent downloads.

Author: encukou

Tools/unicode/makeunicodedata.py

@@ -816,10 +816,31 @@ def makeunicodename(unicode, trace):
 def makestringprep():
     FILE = "Lib/stringprep.py"
 
+    RFC_LOCAL = os.path.join(DATA_DIR, "rfc3454.txt")
+    RFC_URL = "https://www.rfc-editor.org/rfc/rfc3454.txt"
+
     print("--- Preparing", FILE, "...")
 
+    # mkstringprep expects a local copy of RFC 3454. Download it now.
+    # (stringprep is used for URL handling, and if it's not matched
+    # with the compiled unicodedata, downloads would fail.)
+    if not os.path.exists(RFC_LOCAL):
+        download_data(RFC_LOCAL, RFC_URL)
+
     MKSTRINGPREP = "Tools/unicode/mkstringprep.py"
 
+    # mkstringprep needs to be run with a Python version that has "its"
+    # unicode data, since it uses str.lower() and similar.
+    import unicodedata
+    if unicodedata.unidata_version != UNIDATA_VERSION:
+        print()
+        print("!! Skipping mkstringprep -- mimatched Unicode version !!")
+        print()
+        print("Please compile CPython with the updated Unicode database,")
+        print("then use that interpreter to run:")
+        print(f"    python {MKSTRINGPREP} > {FILE}")
+        return
+
     with open(FILE, "w") as f:
         f.truncate()
         subprocess.check_call([sys.executable, MKSTRINGPREP], stdout=f)
@@ -934,21 +955,26 @@ class Difference(Exception):pass
 def open_data(template, version):
     local = os.path.join(DATA_DIR, template % ('-'+version,))
     if not os.path.exists(local):
-        import urllib.request
         if version == '3.2.0':
             # irregular url structure
             url = ('https://www.unicode.org/Public/3.2-Update/'+template) % ('-'+version,)
         else:
             url = ('https://www.unicode.org/Public/%s/ucd/'+template) % (version, '')
-        os.makedirs(os.path.dirname(local), exist_ok=True)
-        urllib.request.urlretrieve(url, filename=local)
+        download_data(local, url)
     if local.endswith('.txt'):
         return open(local, encoding='utf-8')
     else:
         # Unihan.zip
         return open(local, 'rb')
 
 
+def download_data(local, url):
+    import urllib.request
+    os.makedirs(os.path.dirname(local), exist_ok=True)
+    print(f'Downloading {url} to {local}')
+    urllib.request.urlretrieve(url, filename=local)
+
+
 def expand_range(char_range: str) -> Iterator[int]:
     '''
     Parses ranges of code points, as described in UAX #44:

Tools/unicode/mkstringprep.py

@@ -3,9 +3,6 @@
 import unicodedata as unicodedata_current
 from unicodedata import ucd_3_2_0 as unicodedata_320
 
-FILENAME = "Tools/unicode/data/rfc3454.txt"
-URL = "https://www.rfc-editor.org/rfc/rfc3454.txt"
-
 def gen_category(cats):
     for i in range(0, 0x110000):
         if unicodedata_320.category(chr(i)) in cats:
@@ -52,15 +49,7 @@ def compact_set(l):
 
 ############## Read the tables in the RFC #######################
 
-try:
-    data_file = open(FILENAME, encoding='utf-8')
-except FileNotFoundError:
-    import urllib.request
-    os.makedirs(os.path.dirname(FILENAME), exist_ok=True)
-    urllib.request.urlretrieve(URL, filename=FILENAME)
-    data_file = open(FILENAME, encoding='utf-8')
-
-with data_file:
+with open("Tools/unicode/data/rfc3454.txt", encoding='utf-8') as data_file:
     data = data_file.readlines()
 
 tables = []