15349103000

Committed 30 May 2025 02:33PM UTC coverage: 28.052% (-30.9%) from 58.951%

Build # 15349103000

Build Type

push

github

Committed by

antonvw

Commit Message

fixing the segfault

Run Details

7896 of 31160 branches covered (25.34%)

Branch coverage included in aggregate %.

6430 of 19909 relevant lines covered (32.3%)

64.4 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

0.0

/src/factory/unified-diff.cpp

////////////////////////////////////////////////////////////////////////////////
// Name:      unified-diff.cpp
// Purpose:   Implementation of class wex::factory::unified_diff
//            https://www.gnu.org/software/diffutils/manual/html_node/Detailed-Unified.html
// Author:    Anton van Wezenbeek
// Copyright: (c) 2024-2025 Anton van Wezenbeek
////////////////////////////////////////////////////////////////////////////////

#include <boost/tokenizer.hpp>
#include <wex/core/log.h>
#include <wex/core/regex.h>
#include <wex/factory/frame.h>
#include <wex/factory/unified-diff.h>

#include <iostream>

#define NEXT_TOKEN                                                             \
  if (++tok_iter == tokens.end())                                              \
  {                                                                            \
    return false;                                                              \
  }

#define CHANGES_LINES(RANGE, TEXT)                                             \
  for (int i = 0; i < m_range[RANGE]; i++)                                     \
  {                                                                            \
    NEXT_TOKEN                                                                 \
    m_text[TEXT].push_back((*tok_iter).substr(1));                             \
  }

#define HEADER_LINES(REGEX, INTO)                                              \
  if (!parse_header(REGEX, *tok_iter, INTO))                                   \
  {                                                                            \
    return false;                                                              \
  }                                                                            \
  NEXT_TOKEN

#define SKIP_LINES                                                             \
  while (tok_iter != tokens.end())                                             \
  {                                                                            \
    NEXT_TOKEN                                                                 \
    if (!tok_iter->starts_with("diff ") && !tok_iter->starts_with("index "))   \
    {                                                                          \
      break;                                                                   \
    }                                                                          \
  }

namespace wex
{
size_t stoi(const std::string& i)
{
  return i.empty() ? 1 : std::stoi(i);
}
} // namespace wex

wex::factory::unified_diff::unified_diff(const std::string& input)
  : m_input(input)
{
  m_range.fill({0});
}

bool wex::factory::unified_diff::parse()
{
  typedef boost::tokenizer<boost::char_separator<char>> tokenizer;

  tokenizer tokens(m_input, boost::char_separator<char>("\r\n"));

  tokenizer::iterator tok_iter = tokens.begin();

  m_diffs = 0;

  while (tok_iter != tokens.end())
  {
    // skip first lines
    SKIP_LINES;

    // The unified output format starts with a two-line header
    HEADER_LINES("--- a/(.*)", m_path[0]);
    HEADER_LINES("\\+\\+\\+ b/(.*)", m_path[1]);

    m_is_first = true;
    m_is_last  = false;

    // Next come one or more chunks of differences
    while (tok_iter != tokens.end())
    {
      regex r_chunk("@@ -([0-9]+),?([0-9]*) \\+([0-9]+),?([0-9]*) @@.*");

      if (r_chunk.match(*tok_iter) != 4)
      {
        log("unified_diff") << *tok_iter << r_chunk.size();
        return false;
      }

      m_range[0] = wex::stoi(r_chunk[0]);
      m_range[1] = wex::stoi(r_chunk[1]);
      m_range[2] = wex::stoi(r_chunk[2]);
      m_range[3] = wex::stoi(r_chunk[3]);

      m_text.fill({});

      // Now get all - lines and all + lines, collect them, and invoke callback.
      CHANGES_LINES(1, 0);
      CHANGES_LINES(3, 1);

      if (!report_diff())
      {
        return false;
      }

      m_is_first = false;

      if (++tok_iter != tokens.end() && !(*tok_iter).starts_with("@@"))
      {
        m_is_last = true;
        report_diff();
        break; // this was last chunk, continue with header lines
      }
    }
  }

  m_is_last = true;

  report_diff();
  report_diff_finish();

  return true;
}

bool wex::factory::unified_diff::parse_header(
  const std::string& r,
  const std::string& line,
  path&              p)
{
  regex re(r);

  if (!re.match(line))
  {
    log("unified_diff") << line << re.match_data().text();
    return false;
  }

  p = path(re[0]);

  return true;
}

1	////////////////////////////////////////////////////////////////////////////////
2	// Name: unified-diff.cpp
3	// Purpose: Implementation of class wex::factory::unified_diff
4	// https://www.gnu.org/software/diffutils/manual/html_node/Detailed-Unified.html
5	// Author: Anton van Wezenbeek
6	// Copyright: (c) 2024-2025 Anton van Wezenbeek
7	////////////////////////////////////////////////////////////////////////////////
8
9	#include <boost/tokenizer.hpp>
10	#include <wex/core/log.h>
11	#include <wex/core/regex.h>
12	#include <wex/factory/frame.h>
13	#include <wex/factory/unified-diff.h>
14
15	#include <iostream>
16
17	#define NEXT_TOKEN \
18	if (++tok_iter == tokens.end()) \
19	{ \
20	return false; \
21	}
22
23	#define CHANGES_LINES(RANGE, TEXT) \
24	for (int i = 0; i < m_range[RANGE]; i++) \
25	{ \
26	NEXT_TOKEN \
27	m_text[TEXT].push_back((*tok_iter).substr(1)); \
28	}
29
30	#define HEADER_LINES(REGEX, INTO) \
31	if (!parse_header(REGEX, *tok_iter, INTO)) \
32	{ \
33	return false; \
34	} \
35	NEXT_TOKEN
36
37	#define SKIP_LINES \
38	while (tok_iter != tokens.end()) \
39	{ \
40	NEXT_TOKEN \
41	if (!tok_iter->starts_with("diff ") && !tok_iter->starts_with("index ")) \
42	{ \
43	break; \
44	} \
45	}
46
47	namespace wex
48	{
49	size_t stoi(const std::string& i)	×
50	{
51	return i.empty() ? 1 : std::stoi(i);	×
52	}
53	} // namespace wex
54
55	wex::factory::unified_diff::unified_diff(const std::string& input)	×
56	: m_input(input)	×
57	{
58	m_range.fill({0});	×
59	}	×
60
61	bool wex::factory::unified_diff::parse()	×
62	{
63	typedef boost::tokenizer<boost::char_separator<char>> tokenizer;
64
65	tokenizer tokens(m_input, boost::char_separator<char>("\r\n"));	×
66
67	tokenizer::iterator tok_iter = tokens.begin();	×
68
69	m_diffs = 0;	×
70
71	while (tok_iter != tokens.end())	×
72	{
73	// skip first lines
74	SKIP_LINES;	×
75
76	// The unified output format starts with a two-line header
77	HEADER_LINES("--- a/(.*)", m_path[0]);	×
78	HEADER_LINES("\\+\\+\\+ b/(.*)", m_path[1]);	×
79
80	m_is_first = true;	×
81	m_is_last = false;	×
82
83	// Next come one or more chunks of differences
84	while (tok_iter != tokens.end())	×
85	{
86	regex r_chunk("@@ -([0-9]+),?([0-9]) \\+([0-9]+),?([0-9]) @@.*");	×
87
88	if (r_chunk.match(*tok_iter) != 4)	×
89	{
90	log("unified_diff") << *tok_iter << r_chunk.size();	×
91	return false;	×
92	}
93
94	m_range[0] = wex::stoi(r_chunk[0]);	×
95	m_range[1] = wex::stoi(r_chunk[1]);	×
96	m_range[2] = wex::stoi(r_chunk[2]);	×
97	m_range[3] = wex::stoi(r_chunk[3]);	×
98
99	m_text.fill({});	×
100
101	// Now get all - lines and all + lines, collect them, and invoke callback.
102	CHANGES_LINES(1, 0);	×
103	CHANGES_LINES(3, 1);	×
104
105	if (!report_diff())	×
106	{
107	return false;	×
108	}
109
110	m_is_first = false;	×
111
112	if (++tok_iter != tokens.end() && !(*tok_iter).starts_with("@@"))	×
113	{
114	m_is_last = true;	×
115	report_diff();	×
116	break; // this was last chunk, continue with header lines	×
117	}
118	}	×
119	}
120
121	m_is_last = true;	×
122
123	report_diff();	×
124	report_diff_finish();	×
125
126	return true;	×
127	}	×
128
129	bool wex::factory::unified_diff::parse_header(	×
130	const std::string& r,
131	const std::string& line,
132	path& p)
133	{
134	regex re(r);	×
135
136	if (!re.match(line))	×
137	{
138	log("unified_diff") << line << re.match_data().text();	×
139	return false;	×
140	}
141
142	p = path(re[0]);	×
143
144	return true;	×
145	}	×

antonvw / wex / 15349103000

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous