[mwm.py] Review fixes and python3 support

3 files changed, 53 insertions, 35 deletions

diff --git a/tools/python/mwm/dump_mwm.py b/tools/python/mwm/dump_mwm.py
index 251c5cd03c..dfa0203aa2 100755
--- a/tools/python/mwm/dump_mwm.py
+++ b/tools/python/mwm/dump_mwm.py
@@ -4,31 +4,33 @@ import json
 from mwm import MWM
 
 if len(sys.argv) < 2:
-    print 'Dumps some MWM structures.'
-    print 'Usage: {0} <country.mwm>'.format(sys.argv[0])
+    print('Dumps some MWM structures.')
+    print('Usage: {0} <country.mwm>'.format(sys.argv[0]))
     sys.exit(1)
 
 mwm = MWM(open(sys.argv[1], 'rb'))
 mwm.read_types(os.path.join(os.path.dirname(sys.argv[0]), '..', '..', '..', 'data', 'types.txt'))
-print 'Tags:'
-tvv = sorted([(k, v[0], v[1]) for k, v in mwm.tags.iteritems()], key=lambda x: x[1])
+print('Tags:')
+tvv = sorted([(k, v[0], v[1]) for k, v in mwm.tags.items()], key=lambda x: x[1])
 for tv in tvv:
-    print '  {0:<8}: offs {1:9} len {2:8}'.format(tv[0], tv[1], tv[2])
-print 'Version:', mwm.read_version()
-print 'Header:', mwm.read_header()
-print 'Metadata count:', len(mwm.read_metadata())
+    print('  {0:<8}: offs {1:9} len {2:8}'.format(tv[0], tv[1], tv[2]))
+v = mwm.read_version()
+print('Format: {0}, version: {1}'.format(v['fmt'], v['version'].strftime('%Y-%m-%d %H:%M')))
+print('Header: {0}'.format(mwm.read_header()))
+print('Metadata count: {0}'.format(len(mwm.read_metadata())))
 
 cross = mwm.read_crossmwm()
 if cross:
-    print 'Outgoing points:', len(cross['out']), 'incoming:', len(cross['in'])
-    print 'Outgoing regions:', set(cross['neighbours'])
+    print('Outgoing points: {0}, incoming: {1}'.format(len(cross['out']), len(cross['in'])))
+    print('Outgoing regions: {0}'.format(set(cross['neighbours'])))
 
-print 'Sample features:'
+print('Sample features:')
+# Print 5 random features ~10000 features apart
 count = 5
-probability = 1.0 / 1000
+probability = 1.0 / 10000
 for feature in mwm.iter_features():
     if random.random() < probability:
-        print json.dumps(feature, ensure_ascii=False)
+        print(json.dumps(feature, ensure_ascii=False))
         count -= 1
         if count <= 0:
             break
diff --git a/tools/python/mwm/find_feature.py b/tools/python/mwm/find_feature.py
index 4071e7dc08..87b88066d5 100755
--- a/tools/python/mwm/find_feature.py
+++ b/tools/python/mwm/find_feature.py
@@ -3,13 +3,17 @@ import sys, os.path, json
 from mwm import MWM
 
 if len(sys.argv) < 4:
-    print 'Finds features in an mwm file'
-    print 'Usage: {0} <country.mwm> <type> <string>'.format(sys.argv[0])
-    print 'Type: t for inside types, et for exact type, n for names'
+    print('Finds features in an mwm file based on a query')
+    print('Usage: {0} <country.mwm> <type> <string>'.format(sys.argv[0]))
+    print('')
+    print('Type:')
+    print('  t for inside types ("t hwtag" will find all hwtags-*)')
+    print('  et for exact type ("et shop" won\'t find shop-chemist)')
+    print('  n for names, case-sensitive ("n Starbucks" for all starbucks)')
     sys.exit(1)
 
 typ = sys.argv[2].lower()
-find = sys.argv[3]
+find = sys.argv[3].decode('utf-8')
 
 mwm = MWM(open(sys.argv[1], 'rb'))
 mwm.read_header()
@@ -27,4 +31,4 @@ for feature in mwm.iter_features():
             elif typ == 't' and find in t:
                 found = True
     if found:
-        print json.dumps(feature, ensure_ascii=False)
+        print(json.dumps(feature, ensure_ascii=False))
diff --git a/tools/python/mwm/mwm.py b/tools/python/mwm/mwm.py
index abc3552382..99addccf8d 100644
--- a/tools/python/mwm/mwm.py
+++ b/tools/python/mwm/mwm.py
@@ -46,7 +46,7 @@ class MWM:
         cnt = self.read_varuint()
         self.tags = {}
         for i in range(cnt):
-            name = self.read_string(True)
+            name = self.read_string(plain=True)
             offset = self.read_varuint()
             length = self.read_varuint()
             self.tags[name] = (offset, length)
@@ -57,13 +57,13 @@ class MWM:
     def seek_tag(self, tag):
         self.f.seek(self.tags[tag][0])
 
+    def tag_offset(self, tag):
+        return self.f.tell() - self.tags[tag][0]
+
     def inside_tag(self, tag):
-        pos = self.tag_position(tag)
+        pos = self.tag_offset(tag)
         return pos >= 0 and pos < self.tags[tag][1]
 
-    def tag_position(self, tag):
-        return self.f.tell() - self.tags[tag][0]
-
     def read_version(self):
         """Reads 'version' section."""
         self.seek_tag('version')
@@ -128,7 +128,7 @@ class MWM:
         self.seek_tag('meta')
         metadatar = {}
         while self.inside_tag('meta'):
-            tag_pos = self.tag_position('meta')
+            tag_pos = self.tag_offset('meta')
             fields = {}
             if fmt >= 8:
                 sz = self.read_varuint()
@@ -144,7 +144,7 @@ class MWM:
                     t = t & 0x7f
                     t = self.metadata[t] if t < len(self.metadata) else str(t)
                     l = self.read_uint(1)
-                    fields[t] = self.f.read(l)
+                    fields[t] = self.f.read(l).decode('utf-8')
                     if is_last:
                         break
 
@@ -189,7 +189,7 @@ class MWM:
         neighbours = []
         for i in range(neighboursCount):
             size = self.read_uint(4)
-            neighbours.append(self.f.read(size))
+            neighbours.append(self.f.read(size).decode('utf-8'))
         return { 'in': incoming, 'out': outgoing, 'matrix': matrix, 'neighbours': neighbours }
 
     class GeomType:
@@ -314,9 +314,13 @@ class MWM:
             b = self.f.read(1)
             if not b:
                 return res
-            res |= (ord(b[0]) & 0x7F) << shift
+            try:
+                bc = ord(b)
+            except TypeError:
+                bc = b
+            res |= (bc & 0x7F) << shift
             shift += 7
-            more = ord(b[0]) >= 0x80
+            more = bc >= 0x80
         return res
 
     def zigzag_decode(self, uint):
@@ -353,6 +357,7 @@ class MWM:
         return self.mwm_decode_delta(u, ref)
 
     def to_4326(self, point):
+        """Convert a point in maps.me-mercator CS to WGS-84 (EPSG:4326)."""
         if self.coord_size is None:
             raise Exception('Call read_header() first.')
         merc_bounds = (-180.0, -180.0, 180.0, 180.0)  # Xmin, Ymin, Xmax, Ymax
@@ -374,9 +379,10 @@ class MWM:
         pmax = self.to_4326(rmax)
         return (pmin[0], pmin[1], pmax[0], pmax[1])
 
-    def read_string(self, plain=False):
+    def read_string(self, plain=False, decode=True):
         length = self.read_varuint() + (0 if plain else 1)
-        return self.f.read(length)
+        s = self.f.read(length)
+        return s.decode('utf-8') if decode else s
 
     def read_uint_array(self):
         length = self.read_varuint()
@@ -390,13 +396,16 @@ class MWM:
         if sz & 1 != 0:
             return str(sz >> 1)
         sz = (sz >> 1) + 1
-        return self.f.read(sz)
+        return self.f.read(sz).decode('utf-8')
 
     def read_multilang(self):
         def find_multilang_next(s, i):
             i += 1
             while i < len(s):
-                c = struct.unpack('B', s[i])[0]
+                try:
+                    c = ord(s[i])
+                except:
+                    c = s[i]
                 if c & 0xC0 == 0x80:
                     break
                 if c & 0x80 == 0:
@@ -416,13 +425,16 @@ class MWM:
                 i += 1
             return i
 
-        s = self.read_string()
+        s = self.read_string(decode=False)
         langs = {}
         i = 0
         while i < len(s):
             n = find_multilang_next(s, i)
-            lng = struct.unpack('B', s[i])[0] & 0x3F
+            try:
+                lng = ord(s[i]) & 0x3F
+            except TypeError:
+                lng = s[i] & 0x3F
             if lng < len(self.languages):
-                langs[self.languages[lng]] = s[i+1:n]
+                langs[self.languages[lng]] = s[i+1:n].decode('utf-8')
             i = n
         return langs