add component statistics generation

build.py

@@ -35,13 +35,13 @@ def parse(file):
     data['category'] = data['filename'].str.split('/', 3, expand=True)[2]
     data['lang'] = data['filename'].str.split('/', 4, expand=True)[3]
 
-    # TODO: build global.csv and progress.csv
     build_index_adoc(data)
 
     build_language_adoc(data)
     build_language_stats(data)
 
-    # TODO: build each language.csv
+    build_component_adoc(data)
+    build_component_stats(data)
 
     # TODO: build each component.adoc
 
@@ -140,8 +140,6 @@ include::example$language/__LANG__.csv[]
 
 def build_language_stats(data):
     """Build csv files used in adoc file per lang"""
-    #
-
     #
     # Generate progress.csv
     #
@@ -177,6 +175,82 @@ def build_language_stats(data):
         out = out[["component", "pageCount", "pageOver80", "totalMessage", "translatedMessages", "totalSourceWords", "translatedSourceWords", "wordsProgress" ]]
         out.to_csv("modules/ROOT/examples/language/"+lang+".csv", header=headers,index=False)
 
+def build_component_adoc(data):
+    """Build one adoc file per component"""
+
+    hop = """= __COMPONENT__
+
+This component contains:
+
+* __NB_PAGES__ pages
+* __NB_MESSAGES__ messages to translate
+* __NB_WORDS__ words to translate
+
+You can probably.footnote:[All component aren't available for translation for now
+because our translation platform is not fully deployed. Please ask the translation
+mailing list trans@lists.fedoraproject.org if a missing component bother you.]
+translate this component in: https://translate.stg.fedoraproject.org/projects/
+
+[%header,format=csv]
+.Statistics per component
+|===
+include::example$component/__COMPONENT__.csv[]
+|===
+
+"""
+
+    data = data[data.category == "pot"]
+
+    for component in data.component.unique():
+        data_component = data[data.component == component]
+        nb_pages = len(data_component.filename.unique())
+        nb_messages = data_component.totalMessage.sum()
+        nb_words = data_component.totalSourceWords.sum()
+
+        with open("modules/ROOT/pages/component/"+component+".adoc", "w") as f:
+            to_write = hop.replace('__COMPONENT__', component)
+            to_write = to_write.replace('__NB_PAGES__', str(nb_pages))
+            to_write = to_write.replace('__NB_MESSAGES__', str(nb_messages))
+            to_write = to_write.replace('__NB_WORDS__', str(nb_words))
+            f.write(to_write)
+
+def build_component_stats(data):
+    """Build csv files used in adoc file per component"""
+    #
+    # Generate progress.csv
+    #
+    details = data[data.category == "po"]
+
+    for component in details.component.unique():
+        stats = details[details.component == component]
+
+        stats_lang = stats.groupby(["lang"]).sum()
+        stats_lang = stats_lang.reset_index()
+
+        stats_lang = stats_lang[["lang", "translatedSourceWords", "totalSourceWords"]]
+        stats_lang["wordsProgress"] = stats_lang["translatedSourceWords"] / stats_lang["totalSourceWords"]
+        stats_lang["wordsProgress"] = pandas.Series(["{0:.2f}%".format(val * 100) for val in stats_lang["wordsProgress"]], index = stats_lang.index)
+
+
+        # create stats per filename (aka pages)
+        stats_file = stats.groupby(["lang", "filename"]).sum()
+        stats_file["pageprogress"] = stats_file["translatedSourceWords"] / stats_file["totalSourceWords"]
+        stats_file["pageOver80"] = (stats_file["pageprogress"] >= 0.8) *1
+        stats_file["pageCount"] = stats.groupby(["component", "filename", "lang"]).filename.nunique().value_counts().index.tolist()[0]
+        stats_file = stats_file.groupby("lang").sum().reset_index()
+        stats_file = stats_file[["lang", "pageCount", "pageOver80"]]
+
+
+        out = pandas.merge(stats_lang, stats_file, on="lang")
+        out = out.drop_duplicates()
+
+        out['lang'] = pandas.Series(["xref:language/{code}.adoc[{code}]".format(code = val) for val in out["lang"]], index = out.index)
+
+        # write csv
+        headers = ["Language code", "Pages", "Pages >= 80%", "Words progress"]
+        out = out[["lang", "pageCount", "pageOver80", "wordsProgress" ]]
+        out.to_csv("modules/ROOT/examples/component/"+component+".csv", header=headers,index=False)
+
 def store(dataset, name):
     """Store dataset to csv"""
     dataset.to_csv(name, index=False)