We don't need the investigate() function anymore.

[philipp/winterrodeln/wrpylib.git] / wrpylib / wrmwcache.py

diff --git a/wrpylib/wrmwcache.py b/wrpylib/wrmwcache.py
index 6f29e2b04195f26a2957088f25f8a1f33f681cb3..95e0e96805f3d0ba8a5af4ff1e97c23890b372c1 100644 (file)
--- a/wrpylib/wrmwcache.py
+++ b/wrpylib/wrmwcache.py
@@ -1,5 +1,4 @@
-#!/usr/bin/python2.7
-# -*- coding: iso-8859-15 -*-
+#!/usr/bin/python3.4
 # $Id$
 # $HeadURL$
 """Contains functions that maintain/update the cache tables."""
@@ -7,8 +6,7 @@ from sqlalchemy import schema
 from sqlalchemy.sql import select
 from sqlalchemy.sql.expression import func as sqlfunc
 from osgeo import ogr
-import formencode
-from wrpylib import mwdb, wrmwdb, mwmarkup, wrmwmarkup
+from wrpylib import mwdb, wrmwdb, wrmwmarkup, wrvalidators
 
 
 class UpdateCacheError(RuntimeError):
@@ -29,6 +27,8 @@ def update_wrsledruncache(connection):
     categorylinks = mwdb.categorylinks_table(metadata)
     revision = mwdb.revision_table(metadata)
     text = mwdb.text_table(metadata)
+    class Sledrun:
+        pass
 
     transaction = connection.begin()
 
@@ -44,13 +44,15 @@ def update_wrsledruncache(connection):
     
     # Refill wrsledruncache table
     for sledrun_page in sledrun_pages:
-        try: 
-            start, end, sledrun = wrmwmarkup.rodelbahnbox_to_sledrun(sledrun_page.old_text)
+        try:
+            rodelbahnbox = wrvalidators.rodelbahnbox_from_str(sledrun_page.old_text)
+            sledrun = wrmwmarkup.sledrun_from_rodelbahnbox(rodelbahnbox, Sledrun())
             sledrun.page_id = sledrun_page.page_id
             sledrun.page_title = sledrun_page.page_title
-            sledrun.under_construction = connection.execute(select([categorylinks], (categorylinks.c.cl_from==sledrun_page.page_id) & (categorylinks.c.cl_to == 'In_Arbeit')).alias('x').count()).fetchone()[0] > 0 # It would be better to do this in the query above
+            sledrun.name_url = wrvalidators.sledrun_page_title_to_pretty_url(sledrun_page.page_title)
+            sledrun.under_construction = connection.execute(select([sqlfunc.count()], (categorylinks.c.cl_from==sledrun_page.page_id) & (categorylinks.c.cl_to == 'In_Arbeit')).alias('x')).fetchone()[0] > 0
             connection.execute(wrsledruncache.insert(sledrun.__dict__))
-        except (RuntimeError, formencode.Invalid) as e:
+        except ValueError as e:
             transaction.rollback()
             error_msg = "Error at sled run '{0}': {1}".format(sledrun_page.page_title, str(e))
             raise UpdateCacheError(error_msg, sledrun_page.page_title, e)
@@ -71,6 +73,8 @@ def update_wrinncache(connection):
     categorylinks = mwdb.categorylinks_table(metadata)
     revision = mwdb.revision_table(metadata)
     text = mwdb.text_table(metadata)
+    class Inn:
+        pass
 
     transaction = connection.begin()
 
@@ -84,13 +88,14 @@ def update_wrinncache(connection):
         
     # Refill wrinncache table
     for inn_page in inn_pages:
-        try: 
-            start, end, inn = wrmwmarkup.gasthausbox_to_inn(inn_page.old_text)
+        try:
+            gasthausbox = wrvalidators.gasthausbox_from_str(inn_page.old_text)
+            inn = wrmwmarkup.inn_from_gasthausbox(gasthausbox, Inn())
             inn.page_id = inn_page.page_id
             inn.page_title = inn_page.page_title
-            inn.under_construction = connection.execute(select([categorylinks], (categorylinks.c.cl_from==inn_page.page_id) & (categorylinks.c.cl_to == 'In_Arbeit')).alias('x').count()).fetchone()[0] > 0 # It would be better to do this in the query above
+            inn.under_construction = connection.execute(select([sqlfunc.count()], (categorylinks.c.cl_from==inn_page.page_id) & (categorylinks.c.cl_to == 'In_Arbeit')).alias('x')).fetchone()[0] > 0 # It would be better to do this in the query above
             connection.execute(wrinncache.insert(inn.__dict__))
-        except (RuntimeError, formencode.Invalid) as e:
+        except ValueError as e:
             transaction.rollback()
             error_msg = "Error as inn '{0}': {1}".format(inn_page.page_title, str(e))
             raise UpdateCacheError(error_msg, inn_page.page_title, e)
@@ -108,7 +113,6 @@ def update_wrreportcache(connection, page_id=None):
     >>> update_wrreportcache(engine.connect())
     """
     metadata = schema.MetaData()
-    wrreport = wrmwdb.wrreport_table(metadata)
     wrreportcache = wrmwdb.wrreportcache_table(metadata)
     transaction = connection.begin()
 
@@ -168,11 +172,16 @@ def update_wrmapcache(connection):
     
     # Refill wrmappointcache and wrmappathcache tables
     for sledrun_page in sledrun_pages:
-        try: 
-            start, content, endtag, end = mwmarkup.find_tag(sledrun_page.old_text, 'wrmap')
-            if content is None:
-                continue
-            geojson = wrmwmarkup.parse_wrmap(sledrun_page.old_text[start:end])
+        try:
+            import mwparserfromhell
+            wikicode = mwparserfromhell.parse(sledrun_page.old_text)
+            wrmap_list = wikicode.filter_tags(recursive=False, matches=lambda tag: tag.tag == 'wrmap')
+            if len(wrmap_list) == 0:
+                continue  # not wrmap in page
+            if len(wrmap_list) > 1:
+                raise UpdateCacheError('{} <wrmap ...> entries found in article "{}"'.format(len(wrmap_list), sledrun_page.page_title))
+            wrmap = wrmap_list[0]
+            geojson = wrmwmarkup.parse_wrmap(str(wrmap))
 
             for feature in geojson['features']:
                 properties = feature['properties']
@@ -241,4 +250,3 @@ def update_wrregioncache(connection):
 
     # commit
     transaction.commit()
-