strutil.cc revision 415a9b1b3aa75238c9893ea5f117c0877ecc8329
1// Copyright 2015 Google Inc. All rights reserved
2//
3// Licensed under the Apache License, Version 2.0 (the "License");
4// you may not use this file except in compliance with the License.
5// You may obtain a copy of the License at
6//
7//      http://www.apache.org/licenses/LICENSE-2.0
8//
9// Unless required by applicable law or agreed to in writing, software
10// distributed under the License is distributed on an "AS IS" BASIS,
11// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12// See the License for the specific language governing permissions and
13// limitations under the License.
14
15// +build ignore
16
17#include "strutil.h"
18
19#include <ctype.h>
20#include <limits.h>
21#include <unistd.h>
22
23#include <stack>
24#include <utility>
25
26#include "log.h"
27
28WordScanner::Iterator& WordScanner::Iterator::operator++() {
29  int len = static_cast<int>(in->size());
30  for (s = i; s < len; s++) {
31    if (!isspace((*in)[s]))
32      break;
33  }
34  if (s == len) {
35    in = NULL;
36    s = 0;
37    i = 0;
38    return *this;
39  }
40  for (i = s; i < len; i++) {
41    if (isspace((*in)[i]))
42      break;
43  }
44  return *this;
45}
46
47StringPiece WordScanner::Iterator::operator*() const {
48  return in->substr(s, i - s);
49}
50
51WordScanner::WordScanner(StringPiece in)
52    : in_(in) {
53}
54
55WordScanner::Iterator WordScanner::begin() const {
56  Iterator iter;
57  iter.in = &in_;
58  iter.s = 0;
59  iter.i = 0;
60  ++iter;
61  return iter;
62}
63
64WordScanner::Iterator WordScanner::end() const {
65  Iterator iter;
66  iter.in = NULL;
67  iter.s = 0;
68  iter.i = 0;
69  return iter;
70}
71
72void WordScanner::Split(vector<StringPiece>* o) {
73  for (StringPiece t : *this)
74    o->push_back(t);
75}
76
77WordWriter::WordWriter(string* o)
78    : out_(o),
79      needs_space_(false) {
80}
81
82void WordWriter::MaybeAddWhitespace() {
83  if (needs_space_) {
84    out_->push_back(' ');
85  } else {
86    needs_space_ = true;
87  }
88}
89
90void WordWriter::Write(StringPiece s) {
91  MaybeAddWhitespace();
92  AppendString(s, out_);
93}
94
95ScopedTerminator::ScopedTerminator(StringPiece s)
96    : s_(s), c_(s[s.size()]) {
97  const_cast<char*>(s_.data())[s_.size()] = '\0';
98}
99
100ScopedTerminator::~ScopedTerminator() {
101  const_cast<char*>(s_.data())[s_.size()] = c_;
102}
103
104void AppendString(StringPiece str, string* out) {
105  out->append(str.begin(), str.end());
106}
107
108bool HasPrefix(StringPiece str, StringPiece prefix) {
109  ssize_t size_diff = str.size() - prefix.size();
110  return size_diff >= 0 && str.substr(0, prefix.size()) == prefix;
111}
112
113bool HasSuffix(StringPiece str, StringPiece suffix) {
114  ssize_t size_diff = str.size() - suffix.size();
115  return size_diff >= 0 && str.substr(size_diff) == suffix;
116}
117
118StringPiece TrimSuffix(StringPiece str, StringPiece suffix) {
119  ssize_t size_diff = str.size() - suffix.size();
120  if (size_diff < 0 || str.substr(size_diff) != suffix)
121    return str;
122  return str.substr(0, size_diff);
123}
124
125Pattern::Pattern(StringPiece pat)
126    : pat_(pat), percent_index_(pat.find('%')) {
127}
128
129bool Pattern::Match(StringPiece str) const {
130  if (percent_index_ == string::npos)
131    return str == pat_;
132  return MatchImpl(str);
133}
134
135bool Pattern::MatchImpl(StringPiece str) const {
136  return (HasPrefix(str, pat_.substr(0, percent_index_)) &&
137          HasSuffix(str, pat_.substr(percent_index_ + 1)));
138}
139
140void Pattern::AppendSubst(StringPiece str, StringPiece subst,
141                          string* out) const {
142  if (percent_index_ == string::npos) {
143    if (str == pat_) {
144      AppendString(subst, out);
145      return;
146    } else {
147      AppendString(str, out);
148      return;
149    }
150  }
151
152  if (MatchImpl(str)) {
153    size_t subst_percent_index = subst.find('%');
154    if (subst_percent_index == string::npos) {
155      AppendString(subst, out);
156      return;
157    } else {
158      AppendString(subst.substr(0, subst_percent_index), out);
159      AppendString(str.substr(percent_index_,
160                              str.size() - pat_.size() + 1), out);
161      AppendString(subst.substr(subst_percent_index + 1), out);
162      return;
163    }
164  }
165  AppendString(str, out);
166}
167
168void Pattern::AppendSubstRef(StringPiece str, StringPiece subst,
169                             string* out) const {
170  if (percent_index_ != string::npos && subst.find('%') != string::npos) {
171    AppendSubst(str, subst, out);
172    return;
173  }
174  StringPiece s = TrimSuffix(str, pat_);
175  out->append(s.begin(), s.end());
176  out->append(subst.begin(), subst.end());
177}
178
179string NoLineBreak(const string& s) {
180  size_t index = s.find('\n');
181  if (index == string::npos)
182    return s;
183  string r = s;
184  while (index != string::npos) {
185    r = r.substr(0, index) + "\\n" + r.substr(index + 1);
186    index = r.find('\n', index + 2);
187  }
188  return r;
189}
190
191StringPiece TrimLeftSpace(StringPiece s) {
192  size_t i = 0;
193  for (; i < s.size(); i++) {
194    if (isspace(s[i]))
195      continue;
196    char n = s.get(i+1);
197    if (s[i] == '\\' && (n == '\r' || n == '\n')) {
198      i++;
199      continue;
200    }
201    break;
202  }
203  return s.substr(i, s.size() - i);
204}
205
206StringPiece TrimRightSpace(StringPiece s) {
207  size_t i = 0;
208  for (; i < s.size(); i++) {
209    char c = s[s.size() - 1 - i];
210    if (isspace(c)) {
211      if ((c == '\r' || c == '\n') && s.get(s.size() - 2 - i) == '\\')
212        i++;
213      continue;
214    }
215    break;
216  }
217  return s.substr(0, s.size() - i);
218}
219
220StringPiece TrimSpace(StringPiece s) {
221  return TrimRightSpace(TrimLeftSpace(s));
222}
223
224StringPiece Dirname(StringPiece s) {
225  size_t found = s.rfind('/');
226  if (found == string::npos)
227    return StringPiece(".");
228  if (found == 0)
229    return StringPiece("");
230  return s.substr(0, found);
231}
232
233StringPiece Basename(StringPiece s) {
234  size_t found = s.rfind('/');
235  if (found == string::npos || found == 0)
236    return s;
237  return s.substr(found + 1);
238}
239
240StringPiece GetExt(StringPiece s) {
241  size_t found = s.rfind('.');
242  if (found == string::npos)
243    return StringPiece("");
244  return s.substr(found);
245}
246
247StringPiece StripExt(StringPiece s) {
248  size_t slash_index = s.rfind('/');
249  size_t found = s.rfind('.');
250  if (found == string::npos ||
251      (slash_index != string::npos && found < slash_index))
252    return s;
253  return s.substr(0, found);
254}
255
256void NormalizePath(string* o, size_t start_index) {
257  size_t j = start_index;
258  size_t prev_start = start_index;
259  for (size_t i = start_index; i <= o->size(); i++) {
260    char c = (*o)[i];
261    if (c != '/' && c != 0) {
262      (*o)[j] = c;
263      j++;
264      continue;
265    }
266
267    StringPiece prev_dir = StringPiece(o->data() + prev_start, j - prev_start);
268    if (prev_dir == ".") {
269      j--;
270    } else if (prev_dir == "..") {
271      j -= 4;
272      j = o->rfind('/', j);
273      if (j == string::npos) {
274        j = start_index;
275      } else {
276        j++;
277      }
278    } else if (!prev_dir.empty()) {
279      if (c) {
280        (*o)[j] = c;
281        j++;
282      }
283    }
284    prev_start = j;
285  }
286  if (j > 1 && (*o)[j-1] == '/')
287    j--;
288  o->resize(j);
289}
290
291void AbsPath(StringPiece s, string* o) {
292  if (s.get(0) == '/') {
293    o->clear();
294  } else {
295    char buf[PATH_MAX];
296    if (!getcwd(buf, PATH_MAX)) {
297      fprintf(stderr, "getcwd failed\n");
298      CHECK(false);
299    }
300
301    CHECK(buf[0] == '/');
302    *o = buf;
303    *o += '/';
304  }
305  AppendString(s, o);
306  NormalizePath(o, 1);
307}
308
309template<typename Cond>
310size_t FindOutsideParenImpl(StringPiece s, Cond cond) {
311  bool prev_backslash = false;
312  stack<char> paren_stack;
313  for (size_t i = 0; i < s.size(); i++) {
314    char c = s[i];
315    if (cond(c) && paren_stack.empty() && !prev_backslash) {
316      return i;
317    }
318    switch (c) {
319      case '(':
320        paren_stack.push(')');
321        break;
322      case '{':
323        paren_stack.push('}');
324        break;
325
326      case ')':
327      case '}':
328        if (!paren_stack.empty() && c == paren_stack.top()) {
329          paren_stack.pop();
330        }
331        break;
332    }
333    prev_backslash = c == '\\' && !prev_backslash;
334  }
335  return string::npos;
336}
337
338size_t FindOutsideParen(StringPiece s, char c) {
339  return FindOutsideParenImpl(s, [&c](char d){return c == d;});
340}
341
342size_t FindTwoOutsideParen(StringPiece s, char c1, char c2) {
343  return FindOutsideParenImpl(s, [&c1, &c2](char d){
344      return d == c1 || d == c2;
345    });
346}
347
348size_t FindThreeOutsideParen(StringPiece s, char c1, char c2, char c3) {
349  return FindOutsideParenImpl(s, [&c1, &c2, &c3](char d){
350      return d == c1 || d == c2 || d == c3;
351    });
352}
353
354size_t FindEndOfLine(StringPiece s, size_t e, size_t* lf_cnt) {
355  bool prev_backslash = false;
356  for (; e < s.size(); e++) {
357    char c = s[e];
358    if (c == '\\') {
359      prev_backslash = !prev_backslash;
360    } else if (c == '\n') {
361      ++*lf_cnt;
362      if (!prev_backslash) {
363        return e;
364      }
365      prev_backslash = false;
366    } else if (c != '\r') {
367      prev_backslash = false;
368    }
369  }
370  return e;
371}
372
373StringPiece TrimLeadingCurdir(StringPiece s) {
374  while (s.substr(0, 2) == "./")
375    s = s.substr(2);
376  return s;
377}
378