issue(medcat-tutorials): CU-869c87xpy Make sure all the tests run against current state of core lib (#350)

* CU-869c87tt9: Add small bit to workflow to check what noteboo patching does

* CU-869c87xpy: Add same debug output to migration and relcat workflow

* CU-869c87xpy: Fix install target changers

* CU-869c87xpy: Apply notebook patch more robustly (i.e quotes)

* CU-869c87xpy: Fix regex eating the ending quote in some cases

* CU-869c87xpy: Add failure option for notebook patching script

* CU-869c87xpy: Remove debug output from workflow

* CU-869c87xpy: Make workflow fial if/when there is not enough changes in the notebook patcher

* CU-869c87xpy: Remove another debug output and fix more patches

* CU-869c87xpy: [TO REMOVE] Add debug output

* Revert "CU-869c87xpy: [TO REMOVE] Add debug output"

This reverts commit b3eebf9.

Author: mart-r

.github/workflows/medcat-v2-tutorials_main.yml

@@ -26,7 +26,7 @@ jobs:
 
       - name: Update install targets in notebooks
         run: |
-          python .ci/patch_notebook_installs.py .
+          python .ci/patch_notebook_installs.py . --expect-min-changes 4
 
       - name: Install dependencies
         run: |
@@ -69,7 +69,7 @@ jobs:
 
       - name: Update install targets in notebooks
         run: |
-          python .ci/patch_notebook_installs.py .
+          python .ci/patch_notebook_installs.py . --expect-min-changes 4
 
       - name: Install dependencies
         run: |

medcat-v2-tutorials/.ci/patch_notebook_installs.py

@@ -2,60 +2,80 @@
 import pathlib
 import re
 from functools import partial
+import argparse
 
 
-# rel_install_path = "../medcat-v2/"
-# abs_install_path = str(pathlib.Path(rel_install_path).resolve())
+rel_install_path = "../medcat-v2/"
+abs_install_path = str(pathlib.Path(rel_install_path).resolve())
 
 # Matches either:
-# 1. `! pip install medcat[extras]`
+# 1. `! pip install medcat[extras]~=version`
 # 2. `! pip install medcat[extras] @ git+...`
 shell_pattern = re.compile(
-    r'(!\s*pip\s+install\s+)(\\["\']?)medcat(\[.*?\])'
-    r'(\s*@\s*git\+[^"\'\s]+)?\2'
+    r'(!\s*pip\s+install\s+)'       # group 1: the install command
+    r'(\\?"?)'                       # group 2: optional opening \"
+    r'medcat'
+    r'(\[.*?\])?'                    # group 3: optional extras
+    r'(?:'
+        r'\s*@\s*git\+[^"\'\s]+'
+        r'|'
+        r'\s*[~=!<>][^"\'\\s]*'
+    r')'
+    # only match \" (escaped quote), never a bare "
+    r'(\\")?'        # group 4: optional closing \"
 )
 req_txt_pattern = re.compile(
     r'^(medcat(\[.*?\])?)\s*@\s*git\+\S+', flags=re.MULTILINE
 )
 
 
 def repl_nb(m, file_path: pathlib.Path):
-    # extras = m[3]
-    old_url = m[4]
-    if old_url and "medcat/v" in old_url:
-        print(f"[WARN] {file_path} refers to alpha/tagged release: "
-              f"{old_url.strip()}")
-    # to_write = f'{m[1]}\\"{abs_install_path}{extras}\\"'
-    to_write = '! pip install \\"pip\\"'
+    extras = m[3] or ""
+    to_write = f'! pip install \\"{abs_install_path}{extras}\\"'
     print(f"[PATCHED] {file_path}\n with: '{to_write}'")
     return to_write
 
 
 def do_patch(nb_path: pathlib.Path,
-             regex: re.Pattern = shell_pattern, repl_method=repl_nb):
+             regex: re.Pattern = shell_pattern,
+             repl_method=repl_nb) -> bool:
     nb_text = nb_path.read_text(encoding="utf-8")
 
     repl = partial(repl_method, file_path=nb_path)
     new_text = regex.sub(repl, nb_text)
 
     if nb_text != new_text:
         nb_path.write_text(new_text, encoding="utf-8")
+        return True
+    return False
 
 
-def main(path: str):
+def main(path: str, expect_min_changes: int):
+    total_changes = 0
     for nb_path in pathlib.Path(path).rglob("**/*.ipynb"):
-        do_patch(nb_path)
+        if do_patch(nb_path):
+            total_changes += 1
+    if expect_min_changes >= 0 and total_changes < expect_min_changes:
+        print(f"Expected a minimum of {expect_min_changes} changes,"
+              f"but only found {total_changes} changes. "
+              "This will force a non-zero exit status so GHA workflow "
+              "can fail")
+        sys.exit(1)
 
 
 if __name__ == "__main__":
-    if len(sys.argv) != 2:
-        print("Usage: python patch_notebook_installs.py <path>")
-        sys.exit(1)
+    parser = argparse.ArgumentParser()
+    parser.add_argument("path", help="The path to start looking at",
+                        type=str)
+    parser.add_argument("--expect-min-changes", "-c",
+                        help="Expect at lest this number of chagnes",
+                        type=int, default=-1)
+    args = parser.parse_args()
 
-    path = sys.argv[1]
+    path = args.path
 
     if not pathlib.Path(path).exists():
         print(f"Path {path} does not exist.")
         sys.exit(1)
 
-    main(path)
+    main(path, args.expect_min_changes)