record whether repo is official or not

kwalitee/cli.py

@@ -21,7 +21,11 @@ class CliCtx():
     def __init__(self, config_file):
         with open(config_file, 'r') as f_config_file:
             self.config = safe_load(f_config_file.read())
-            self.repos = [Repo(self.config, url) for url in self.config['repolist']]
+            self.repos = []
+            for repo_desc in self.config['repolist']:
+                url = repo_desc['url']
+                official = repo_desc.get('official', False)
+                self.repos.append(Repo(self.config, url, official))
 pass_ctx = click.make_pass_decorator(CliCtx)
 
 @click.group()

kwalitee/config.yml

@@ -2,19 +2,26 @@
 repodir: /data/ocrd_all
 # repos to clone and process
 repolist:
-    - https://github.com/ASVLeipzig/cor-asv-ann
-    - https://github.com/ASVLeipzig/cor-asv-fst
-    - https://github.com/OCR-D/ocrd_calamari
-    - https://github.com/OCR-D/ocrd_im6convert
-    - https://github.com/OCR-D/ocrd_keraslm
-    - https://github.com/OCR-D/ocrd_kraken
-    - https://github.com/OCR-D/ocrd_ocropy
-    - https://github.com/OCR-D/ocrd_olena
-    - https://github.com/OCR-D/ocrd_segment
-    - https://github.com/OCR-D/ocrd_tesserocr
-    - https://github.com/cisocrgroup/ocrd_cis
-    - https://github.com/mjenckel/LAYoutERkennung
-    - https://github.com/ocr-d-modul-2-segmentierung/segmentation-runner
-    - https://github.com/qurator-spk/dinglehopper
-    - https://github.com/qurator-spk/pixelwise_segmentation_SBB
-    - https://github.com/seuretm/ocrd_typegroups_classifier
+  - url: https://github.com/ASVLeipzig/cor-asv-ann
+    official: true
+  - url: https://github.com/ASVLeipzig/cor-asv-fst
+    official: true
+  - url: https://github.com/OCR-D/ocrd_calamari
+  - url: https://github.com/OCR-D/ocrd_im6convert
+  - url: https://github.com/OCR-D/ocrd_keraslm
+  - url: https://github.com/OCR-D/ocrd_kraken
+  - url: https://github.com/OCR-D/ocrd_ocropy
+  - url: https://github.com/OCR-D/ocrd_olena
+  - url: https://github.com/OCR-D/ocrd_segment
+  - url: https://github.com/OCR-D/ocrd_tesserocr
+    official: true
+  - url: https://github.com/cisocrgroup/ocrd_cis
+    official: true
+  - url: https://github.com/mjenckel/LAYoutERkennung
+    official: true
+  - url: https://github.com/ocr-d-modul-2-segmentierung/segmentation-runner
+    official: true
+  - url: https://github.com/qurator-spk/dinglehopper
+  - url: https://github.com/qurator-spk/pixelwise_segmentation_SBB
+  - url: https://github.com/seuretm/ocrd_typegroups_classifier
+    official: true

kwalitee/repo.py

@@ -10,10 +10,11 @@ LOG = getLogger('kwalitee.repo')
 
 class Repo():
 
-    def __init__(self, config, url):
+    def __init__(self, config, url, official=False):
         self.url = url
         self.config = config
         self.name = Path(url).name
+        self.official = official
         self.path = Path(self.config['repodir'], self.name)
 
     def __str__(self):
@@ -73,6 +74,7 @@ class Repo():
     def to_json(self):
         desc = {}
         desc['url'] = self.url
+        desc['official'] = self.official
         desc['org_plus_name'] = '/'.join(self.url.split('/')[-2:])
         desc['name'] = self.name
         desc['files'] = self.get_file_contents()

requirements.txt

 click >=7
-ocrd >= 1.0.0
+ocrd >= 2.0.0a1
 pyyaml
 requests