Merge pull request OCA#204 from StefanRijnhart/5.0-backport_records

[PRT] Backport improvements to the records module from 8.0

.gitignore

@@ -0,0 +1,2 @@
+*.pyc
+*~

bin/addons/openupgrade_records/__init__.py

@@ -1,2 +1 @@
 import model
-import lib

bin/addons/openupgrade_records/__openerp__.py

@@ -2,7 +2,7 @@
 ##############################################################################
 #
 #    OpenERP, Open Source Management Solution
-#    This module Copyright (C) 2012 OpenUpgrade community
+#    This module Copyright (C) 2012-2014 OpenUpgrade community
 #    https://launchpad.net/~openupgrade-committers
 #
 #    Contributors:
@@ -28,7 +28,7 @@
     'name': 'OpenUpgrade Records',
     'version': '0.2',
     'category': 'Normal',
-    'description': """Allow OpenUpgrade records to be 
+    'description': """Allow OpenUpgrade records to be
 stored in the database and compare with other servers.
 
 This module depends on OpenERP client lib:
@@ -57,7 +57,7 @@
     'installable': True,
     'auto_install': False,
     'external_dependencies': {
-        'python' : ['openerplib'],
+        'python': ['openerplib'],
         },
 }
 # vim:expandtab:smartindent:tabstop=4:softtabstop=4:shiftwidth=4:

bin/addons/openupgrade_records/__terp__.py

@@ -2,7 +2,7 @@
 ##############################################################################
 #
 #    OpenERP, Open Source Management Solution
-#    This module Copyright (C) 2012 OpenUpgrade community
+#    This module Copyright (C) 2012-2014 OpenUpgrade community
 #    https://launchpad.net/~openupgrade-committers
 #
 #    Contributors:
@@ -28,7 +28,7 @@
     'name': 'OpenUpgrade Records',
     'version': '0.2',
     'category': 'Normal',
-    'description': """Allow OpenUpgrade records to be 
+    'description': """Allow OpenUpgrade records to be
 stored in the database and compare with other servers.
 
 This module depends on OpenERP client lib:
@@ -57,7 +57,7 @@
     'installable': True,
     'auto_install': False,
     'external_dependencies': {
-        'python' : ['openerplib'],
+        'python': ['openerplib'],
         },
 }
 # vim:expandtab:smartindent:tabstop=4:softtabstop=4:shiftwidth=4:

bin/addons/openupgrade_records/lib/__init__.py

@@ -1,2 +0,0 @@
-import apriori
-import compare

bin/addons/openupgrade_records/lib/compare.py

@@ -24,30 +24,26 @@
 #   layouts from the OpenUpgrade server.
 #####################################################################
 
+import collections
 import copy
 
 try:
     from openerp.addons.openupgrade_records.lib import apriori
 except ImportError:
     from openupgrade_records.lib import apriori
 
-keys = [
-    'module',
-    'mode',
-    'model',
-    'field',
-    'type',
-    'isfunction',
-    'relation',
-    'required',
-    'selection_keys',
-    'req_default',
-    'inherits',
-    ]
 
 def module_map(module):
-    return apriori.renamed_modules.get(
-        module, module)
+    return apriori.renamed_modules.get(module, module)
+
+
+def model_map(model):
+    return apriori.renamed_models.get(model, model)
+
+
+IGNORE_FIELDS = [
+    ]
+
 
 def compare_records(dict_old, dict_new, fields):
     """
@@ -62,37 +58,41 @@ def compare_records(dict_old, dict_new, fields):
             if (module_map(dict_old[field]) != dict_new[field]):
                 return False
         elif field == 'model':
-            if (apriori.renamed_models.get(
-                    dict_old[field], dict_old[field]) != dict_new[field]):
-                return False
-        else:
-            if dict_old[field] != dict_new[field]:
+            if (model_map(dict_old[field]) != dict_new[field]):
                 return False
+        elif dict_old[field] != dict_new[field]:
+            return False
     return True
 
+
 def search(item, item_list, fields):
     """
     Find a match of a dictionary in a list of similar dictionaries
     with respect to the keys in the 'fields' arguments.
     Return the item if found or None.
     """
-    for i in item_list:
-        if not compare_records(item, i, fields):
+    for other in item_list:
+        if not compare_records(item, other, fields):
             continue
-        return i
+        return other
+    # search for renamed fields
+    if 'field' in fields:
+        for other in item_list:
+            if not item['field'] or item['field'] != other.get('oldname'):
+                continue
+            if compare_records(dict(item, field=other['field']), other, fields):
+                return other
     return None
 
+
 def fieldprint(old, new, field, text, reprs):
     fieldrepr = "%s (%s)" % (old['field'], old['type'])
-    repr = '%s / %s / %s' % (
-        old['module'].ljust(12), old['model'].ljust(24), fieldrepr.ljust(30))
-    if text:
-        reprs.setdefault(module_map(old['module']), []).append(
-            "%s: %s" % (repr, text))
-    else:
-        reprs.setdefault(module_map(old['module']), []).append(
-            "%s: %s is now \'%s\' ('%s')" % (
-                repr, field, new[field], old[field]))
+    fullrepr = '%-12s / %-24s / %-30s' % (
+        old['module'], old['model'], fieldrepr)
+    if not text:
+        text = "%s is now '%s' ('%s')" % (field, new[field], old[field])
+    reprs[module_map(old['module'])].append("%s: %s" % (fullrepr, text))
+
 
 def report_generic(new, old, attrs, reprs):
     for attr in attrs:
@@ -109,9 +109,13 @@ def report_generic(new, old, attrs, reprs):
                 else:
                     text = "not a function anymore"
                 fieldprint(old, new, None, text, reprs)
-        else:
-            if old[attr] != new[attr]:
-                fieldprint(old, new, attr, None, reprs)
+        elif attr == 'oldname':
+            if new.get('oldname') == old['field']:
+                text = 'was renamed to %s [nothing to to]' % new['field']
+                fieldprint(old, new, None, text, reprs)
+        elif old[attr] != new[attr]:
+            fieldprint(old, new, attr, None, reprs)
+
 
 def compare_sets(old_records, new_records):
     """
@@ -121,20 +125,20 @@ def compare_sets(old_records, new_records):
     module names as keys. Special case is the 'general' key
     which contains overall remarks and matching statistics.
     """
-    reprs = {'general': []}
+    reprs = collections.defaultdict(list)
 
     for record in old_records + new_records:
         record['matched'] = False
     origlen = len(old_records)
-    new_models = set([ column['model'] for column in new_records ])
-    old_models = set([ column['model'] for column in old_records ])
+    new_models = set([column['model'] for column in new_records])
+    old_models = set([column['model'] for column in old_records])
 
     matched_direct = 0
     matched_other_module = 0
     matched_other_type = 0
     matched_other_name = 0
     in_obsolete_models = 0
-    
+
     obsolete_models = []
     for model in old_models:
         if model not in new_models:
@@ -149,15 +153,15 @@ def compare_sets(old_records, new_records):
     for model in new_models:
         if model not in old_models:
             reprs['general'].append('new model %s' % model)
-    
+
     def match(match_fields, report_fields, warn=False):
         count = 0
         for column in copy.copy(old_records):
             found = search(column, new_records, match_fields)
             if found:
                 if warn:
                     pass
-                    #print "Tentatively"
+                    # print "Tentatively"
                 report_generic(found, column, report_fields, reprs)
                 old_records.remove(column)
                 new_records.remove(found)
@@ -166,20 +170,23 @@ def match(match_fields, report_fields, warn=False):
 
     matched_direct = match(
         ['module', 'mode', 'model', 'field'],
-        ['relation', 'type', 'selection_keys', 'inherits', 'isfunction', 'required'])
+        ['relation', 'type', 'selection_keys', 'inherits',
+         'isfunction', 'required', 'oldname'])
 
     # other module, same type and operation
     matched_other_module = match(
         ['mode', 'model', 'field', 'type'],
-        ['module', 'relation', 'selection_keys', 'inherits', 'isfunction', 'required'])
+        ['module', 'relation', 'selection_keys', 'inherits',
+         'isfunction', 'required', 'oldname'])
 
     # other module, same operation, other type
     matched_other_type = match(
         ['mode', 'model', 'field'],
-        ['relation', 'type', 'selection_keys', 'inherits', 'isfunction', 'required'])
+        ['relation', 'type', 'selection_keys', 'inherits',
+         'isfunction', 'required', 'oldname'])
 
     # fields with other names
-    #matched_other_name = match(
+    # matched_other_name = match(
     # ['module', 'type', 'relation'],
     # ['field', 'relation', 'type', 'selection_keys',
     #   'inherits', 'isfunction', 'required'], warn=True)
@@ -190,51 +197,62 @@ def match(match_fields, report_fields, warn=False):
         ]
     for column in old_records:
         # we do not care about removed function fields
-        if not column['isfunction']:
-            if column['mode'] == 'create':
-                column['mode'] = ''
-            fieldprint(
-                column, None, None, "DEL " + ", ".join(
-                    [k + ': ' + str(column[k]) for k in printkeys if column[k]]
-                    ), reprs)
+        if column['isfunction'] or column['field'] in IGNORE_FIELDS:
+            continue
+        if column['mode'] == 'create':
+            column['mode'] = ''
+        fieldprint(
+            column, None, None, "DEL " + ", ".join(
+                [k + ': ' + str(column[k]) for k in printkeys if column[k]]
+                ), reprs)
 
     for column in new_records:
         # we do not care about newly added function fields
-        if not column['isfunction']:
-            if column['mode'] == 'create':
-                column['mode'] = ''
-            fieldprint(
-                column, None, None, "NEW " + ", ".join(
-                    [k + ': ' + str(column[k]) for k in printkeys if column[k]]
-                    ), reprs)
+        if column['isfunction'] or column['field'] in IGNORE_FIELDS:
+            continue
+        if column['mode'] == 'create':
+            column['mode'] = ''
+        fieldprint(
+            column, None, None, "NEW " + ", ".join(
+                [k + ': ' + str(column[k]) for k in printkeys if column[k]]
+                ), reprs)
 
     for line in [
-        "# %d fields matched," % (origlen - len(old_records)),
-        "# Direct match: %d" % matched_direct,
-        "# Found in other module: %d" % matched_other_module,
-        "# Found with different type: %d" % matched_other_type,
-        "# Found with different name: %d" % matched_other_name,
-        "# In obsolete models: %d" % in_obsolete_models,
-        "# Not matched: %d" % len(old_records),
-        "# New columns: %d" % len(new_records),
-        ]:
+            "# %d fields matched," % (origlen - len(old_records)),
+            "# Direct match: %d" % matched_direct,
+            "# Found in other module: %d" % matched_other_module,
+            "# Found with different type: %d" % matched_other_type,
+            "# Found with different name: %d" % matched_other_name,
+            "# In obsolete models: %d" % in_obsolete_models,
+            "# Not matched: %d" % len(old_records),
+            "# New columns: %d" % len(new_records)
+            ]:
         reprs['general'].append(line)
     return reprs
 
+
 def compare_xml_sets(old_records, new_records):
-    reprs = {}
+    reprs = collections.defaultdict(list)
     match_fields = ['module', 'model', 'name']
     for column in copy.copy(old_records):
         found = search(column, new_records, match_fields)
         if found:
             old_records.remove(column)
             new_records.remove(found)
-    for entry in sorted(
-        old_records, key=lambda k: '%s%s' % (k['model'].ljust(128), k['name'])):
-        reprs.setdefault(module_map(entry['module']), []).append(
-            'deleted xml-id of model %s: %s' % (entry['model'], entry['name']))
-    for entry in sorted(
-        new_records, key=lambda k: '%s%s' % (k['model'].ljust(128), k['name'])):
-        reprs.setdefault(module_map(entry['module']), []).append(
-            'new xml-id of model %s: %s' % (entry['model'], entry['name']))
+
+    for record in old_records:
+        record['old'] = True
+    for record in new_records:
+        record['new'] = True
+
+    sorted_records = sorted(
+        old_records + new_records,
+        key=lambda k: (k['model'], 'old' in k, k['name'])
+    )
+    for entry in sorted_records:
+        if 'old' in entry:
+            content = 'DEL %(model)s: %(name)s' % entry
+        elif 'new' in entry:
+            content = 'NEW %(model)s: %(name)s' % entry
+        reprs[module_map(entry['module'])].append(content)
     return reprs

bin/addons/openupgrade_records/model/__init__.py

@@ -3,4 +3,3 @@
 import analysis_wizard
 import generate_records_wizard
 import install_all_wizard
-