OLD | NEW |
| (Empty) |
1 """Coverage data for Coverage.""" | |
2 | |
3 import os | |
4 | |
5 from coverage.backward import iitems, pickle, sorted # pylint: disable=W0622 | |
6 from coverage.files import PathAliases | |
7 from coverage.misc import file_be_gone | |
8 | |
9 | |
10 class CoverageData(object): | |
11 """Manages collected coverage data, including file storage. | |
12 | |
13 The data file format is a pickled dict, with these keys: | |
14 | |
15 * collector: a string identifying the collecting software | |
16 | |
17 * lines: a dict mapping filenames to sorted lists of line numbers | |
18 executed: | |
19 { 'file1': [17,23,45], 'file2': [1,2,3], ... } | |
20 | |
21 * arcs: a dict mapping filenames to sorted lists of line number pairs: | |
22 { 'file1': [(17,23), (17,25), (25,26)], ... } | |
23 | |
24 """ | |
25 | |
26 def __init__(self, basename=None, collector=None): | |
27 """Create a CoverageData. | |
28 | |
29 `basename` is the name of the file to use for storing data. | |
30 | |
31 `collector` is a string describing the coverage measurement software. | |
32 | |
33 """ | |
34 self.collector = collector or 'unknown' | |
35 | |
36 self.use_file = True | |
37 | |
38 # Construct the filename that will be used for data file storage, if we | |
39 # ever do any file storage. | |
40 self.filename = basename or ".coverage" | |
41 self.filename = os.path.abspath(self.filename) | |
42 | |
43 # A map from canonical Python source file name to a dictionary in | |
44 # which there's an entry for each line number that has been | |
45 # executed: | |
46 # | |
47 # { | |
48 # 'filename1.py': { 12: None, 47: None, ... }, | |
49 # ... | |
50 # } | |
51 # | |
52 self.lines = {} | |
53 | |
54 # A map from canonical Python source file name to a dictionary with an | |
55 # entry for each pair of line numbers forming an arc: | |
56 # | |
57 # { | |
58 # 'filename1.py': { (12,14): None, (47,48): None, ... }, | |
59 # ... | |
60 # } | |
61 # | |
62 self.arcs = {} | |
63 | |
64 def usefile(self, use_file=True): | |
65 """Set whether or not to use a disk file for data.""" | |
66 self.use_file = use_file | |
67 | |
68 def read(self): | |
69 """Read coverage data from the coverage data file (if it exists).""" | |
70 if self.use_file: | |
71 self.lines, self.arcs = self._read_file(self.filename) | |
72 else: | |
73 self.lines, self.arcs = {}, {} | |
74 | |
75 def write(self, suffix=None): | |
76 """Write the collected coverage data to a file. | |
77 | |
78 `suffix` is a suffix to append to the base file name. This can be used | |
79 for multiple or parallel execution, so that many coverage data files | |
80 can exist simultaneously. A dot will be used to join the base name and | |
81 the suffix. | |
82 | |
83 """ | |
84 if self.use_file: | |
85 filename = self.filename | |
86 if suffix: | |
87 filename += "." + suffix | |
88 self.write_file(filename) | |
89 | |
90 def erase(self): | |
91 """Erase the data, both in this object, and from its file storage.""" | |
92 if self.use_file: | |
93 if self.filename: | |
94 file_be_gone(self.filename) | |
95 self.lines = {} | |
96 self.arcs = {} | |
97 | |
98 def line_data(self): | |
99 """Return the map from filenames to lists of line numbers executed.""" | |
100 return dict( | |
101 [(f, sorted(lmap.keys())) for f, lmap in iitems(self.lines)] | |
102 ) | |
103 | |
104 def arc_data(self): | |
105 """Return the map from filenames to lists of line number pairs.""" | |
106 return dict( | |
107 [(f, sorted(amap.keys())) for f, amap in iitems(self.arcs)] | |
108 ) | |
109 | |
110 def write_file(self, filename): | |
111 """Write the coverage data to `filename`.""" | |
112 | |
113 # Create the file data. | |
114 data = {} | |
115 | |
116 data['lines'] = self.line_data() | |
117 arcs = self.arc_data() | |
118 if arcs: | |
119 data['arcs'] = arcs | |
120 | |
121 if self.collector: | |
122 data['collector'] = self.collector | |
123 | |
124 # Write the pickle to the file. | |
125 fdata = open(filename, 'wb') | |
126 try: | |
127 pickle.dump(data, fdata, 2) | |
128 finally: | |
129 fdata.close() | |
130 | |
131 def read_file(self, filename): | |
132 """Read the coverage data from `filename`.""" | |
133 self.lines, self.arcs = self._read_file(filename) | |
134 | |
135 def raw_data(self, filename): | |
136 """Return the raw pickled data from `filename`.""" | |
137 fdata = open(filename, 'rb') | |
138 try: | |
139 data = pickle.load(fdata) | |
140 finally: | |
141 fdata.close() | |
142 return data | |
143 | |
144 def _read_file(self, filename): | |
145 """Return the stored coverage data from the given file. | |
146 | |
147 Returns two values, suitable for assigning to `self.lines` and | |
148 `self.arcs`. | |
149 | |
150 """ | |
151 lines = {} | |
152 arcs = {} | |
153 try: | |
154 data = self.raw_data(filename) | |
155 if isinstance(data, dict): | |
156 # Unpack the 'lines' item. | |
157 lines = dict([ | |
158 (f, dict.fromkeys(linenos, None)) | |
159 for f, linenos in iitems(data.get('lines', {})) | |
160 ]) | |
161 # Unpack the 'arcs' item. | |
162 arcs = dict([ | |
163 (f, dict.fromkeys(arcpairs, None)) | |
164 for f, arcpairs in iitems(data.get('arcs', {})) | |
165 ]) | |
166 except Exception: | |
167 pass | |
168 return lines, arcs | |
169 | |
170 def combine_parallel_data(self, aliases=None): | |
171 """Combine a number of data files together. | |
172 | |
173 Treat `self.filename` as a file prefix, and combine the data from all | |
174 of the data files starting with that prefix plus a dot. | |
175 | |
176 If `aliases` is provided, it's a `PathAliases` object that is used to | |
177 re-map paths to match the local machine's. | |
178 | |
179 """ | |
180 aliases = aliases or PathAliases() | |
181 data_dir, local = os.path.split(self.filename) | |
182 localdot = local + '.' | |
183 for f in os.listdir(data_dir or '.'): | |
184 if f.startswith(localdot): | |
185 full_path = os.path.join(data_dir, f) | |
186 new_lines, new_arcs = self._read_file(full_path) | |
187 for filename, file_data in iitems(new_lines): | |
188 filename = aliases.map(filename) | |
189 self.lines.setdefault(filename, {}).update(file_data) | |
190 for filename, file_data in iitems(new_arcs): | |
191 filename = aliases.map(filename) | |
192 self.arcs.setdefault(filename, {}).update(file_data) | |
193 if f != local: | |
194 os.remove(full_path) | |
195 | |
196 def add_line_data(self, line_data): | |
197 """Add executed line data. | |
198 | |
199 `line_data` is { filename: { lineno: None, ... }, ...} | |
200 | |
201 """ | |
202 for filename, linenos in iitems(line_data): | |
203 self.lines.setdefault(filename, {}).update(linenos) | |
204 | |
205 def add_arc_data(self, arc_data): | |
206 """Add measured arc data. | |
207 | |
208 `arc_data` is { filename: { (l1,l2): None, ... }, ...} | |
209 | |
210 """ | |
211 for filename, arcs in iitems(arc_data): | |
212 self.arcs.setdefault(filename, {}).update(arcs) | |
213 | |
214 def touch_file(self, filename): | |
215 """Ensure that `filename` appears in the data, empty if needed.""" | |
216 self.lines.setdefault(filename, {}) | |
217 | |
218 def measured_files(self): | |
219 """A list of all files that had been measured.""" | |
220 return list(self.lines.keys()) | |
221 | |
222 def executed_lines(self, filename): | |
223 """A map containing all the line numbers executed in `filename`. | |
224 | |
225 If `filename` hasn't been collected at all (because it wasn't executed) | |
226 then return an empty map. | |
227 | |
228 """ | |
229 return self.lines.get(filename) or {} | |
230 | |
231 def executed_arcs(self, filename): | |
232 """A map containing all the arcs executed in `filename`.""" | |
233 return self.arcs.get(filename) or {} | |
234 | |
235 def add_to_hash(self, filename, hasher): | |
236 """Contribute `filename`'s data to the Md5Hash `hasher`.""" | |
237 hasher.update(self.executed_lines(filename)) | |
238 hasher.update(self.executed_arcs(filename)) | |
239 | |
240 def summary(self, fullpath=False): | |
241 """Return a dict summarizing the coverage data. | |
242 | |
243 Keys are based on the filenames, and values are the number of executed | |
244 lines. If `fullpath` is true, then the keys are the full pathnames of | |
245 the files, otherwise they are the basenames of the files. | |
246 | |
247 """ | |
248 summ = {} | |
249 if fullpath: | |
250 filename_fn = lambda f: f | |
251 else: | |
252 filename_fn = os.path.basename | |
253 for filename, lines in iitems(self.lines): | |
254 summ[filename_fn(filename)] = len(lines) | |
255 return summ | |
256 | |
257 def has_arcs(self): | |
258 """Does this data have arcs?""" | |
259 return bool(self.arcs) | |
260 | |
261 | |
262 if __name__ == '__main__': | |
263 # Ad-hoc: show the raw data in a data file. | |
264 import pprint, sys | |
265 covdata = CoverageData() | |
266 if sys.argv[1:]: | |
267 fname = sys.argv[1] | |
268 else: | |
269 fname = covdata.filename | |
270 pprint.pprint(covdata.raw_data(fname)) | |
OLD | NEW |