reproducible: _html_breakages: add a check to detect stale history pages. Consider that currently nothing delete them automatically

2 files changed, 14 insertions, 0 deletions

diff --git a/TODO b/TODO
index 8fa26327..c183345d 100644
--- a/TODO
+++ b/TODO
@@ -157,6 +157,7 @@ properties:
 ** link howto on each coreboot/openwrt/netbsd/freebsd page
 ** pkg sets are still amd64 only atm… (and there is 404 link to the armhf page)
 ** bin/_html_indexes.py: bugs = get_bugs() # this variable should not be global, else merely importing _html_indexes always queries UDD
+** add a mean to automatically delete the history pages
 
 * lesser prio
 ** dashboard:
diff --git a/bin/reproducible_html_breakages.py b/bin/reproducible_html_breakages.py
index 67698f16..029d94a4 100755
--- a/bin/reproducible_html_breakages.py
+++ b/bin/reproducible_html_breakages.py
@@ -244,6 +244,17 @@ def alien_rbpkg():
     return bad_files
 
 
+def alien_history():
+    log.info('running alien_history check...')
+    result = query_db('SELECT DISTINCT name FROM sources')
+    actual_packages = [x[0] for x in result]
+    bad_files = []
+    for f in sorted(os.listdir(HISTORY_PATH)):
+        if f.rsplit('.', 1)[0] not in actual_packages:
+            log.warning('%s should not be there', os.path.join(HISTORY_PATH, f))
+    return bad_files
+
+
 def _gen_section(header, pkgs, entries=None):
     if not pkgs and not entries:
         return ''
@@ -275,6 +286,8 @@ def gen_html():
                          entries=alien_rbpkg())
     html += _gen_section('buildinfo files that should not be there:', None,
                          entries=alien_buildinfo())
+    html += _gen_section('history tables that should not be there:', None,
+                         entries=alien_history())
     # diffoscope report where it shouldn't be
     html += _gen_section('are not marked as unreproducible, but they ' +
                          'have a diffoscope file:', not_unrep_with_dbd_file())