19616694313

Committed 23 Nov 2025 08:15PM UTC coverage: 64.297% (-0.1%) from 64.402%

Build # 19616694313

Build Type

push

github

Committed by

antonvw

Commit Message

added diff type to unified_diff

Run Details

18528 of 31566 branches covered (58.7%)

Branch coverage included in aggregate %.

14666 of 20060 relevant lines covered (73.11%)

1525.53 hits per line

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

89.29

/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>
#include <utility>

#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(std::string input)
  : m_input(std::move(input))
{
  m_range.fill({0});
}

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

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

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

  m_diffs = 0;
  m_type  = diff_t::UNKNOWN;

  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;
    m_type     = (m_type == diff_t::UNKNOWN ? diff_t::FIRST : diff_t::OTHER);

    // 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);

      log::trace("diff parse")
        << std::to_underlying(m_type) << m_path[0].string() << m_diffs;
      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;

  if (m_type != diff_t::UNKNOWN)
  {
    m_type = diff_t::LAST;
    log::trace("diff parse LAST") << m_path[0].string() << " " << m_diffs;
  }

  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	#include <utility>
17
18	#define NEXT_TOKEN \
19	if (++tok_iter == tokens.end()) \
20	{ \
21	return false; \
22	}
23
24	#define CHANGES_LINES(RANGE, TEXT) \
25	for (int i = 0; i < m_range[RANGE]; i++) \
26	{ \
27	NEXT_TOKEN \
28	m_text[TEXT].push_back((*tok_iter).substr(1)); \
29	}
30
31	#define HEADER_LINES(REGEX, INTO) \
32	if (!parse_header(REGEX, *tok_iter, INTO)) \
33	{ \
34	return false; \
35	} \
36	NEXT_TOKEN
37
38	#define SKIP_LINES \
39	while (tok_iter != tokens.end()) \
40	{ \
41	NEXT_TOKEN \
42	if (!tok_iter->starts_with("diff ") && !tok_iter->starts_with("index ")) \
43	{ \
44	break; \
45	} \
46	}
47
48	namespace wex
49	{
50	size_t stoi(const std::string& i)	68✔
51	{
52	return i.empty() ? 1 : std::stoi(i);	68✔
53	}
54	} // namespace wex
55
56	wex::factory::unified_diff::unified_diff(std::string input)	15✔
57	: m_input(std::move(input))	15✔
58	{
59	m_range.fill({0});	15✔
60	}	15✔
61
62	bool wex::factory::unified_diff::parse()	11✔
63	{
64	using tokenizer = boost::tokenizer<boost::char_separator<char>>;
65
66	tokenizer tokens(m_input, boost::char_separator<char>("\r\n"));	11✔
67
68	tokenizer::iterator tok_iter = tokens.begin();	11✔
69
70	m_diffs = 0;	11✔
71	m_type = diff_t::UNKNOWN;	11✔
72
73	while (tok_iter != tokens.end())	34✔
74	{
75	// skip first lines
76	SKIP_LINES;	25!
77
78	// The unified output format starts with a two-line header
79	HEADER_LINES("--- a/(.*)", m_path[0]);	42!
80	HEADER_LINES("\\+\\+\\+ b/(.*)", m_path[1]);	42!
81
82	m_is_first = true;	14✔
83	m_is_last = false;	14✔
84	m_type = (m_type == diff_t::UNKNOWN ? diff_t::FIRST : diff_t::OTHER);	14✔
85
86	// Next come one or more chunks of differences
87	while (tok_iter != tokens.end())	25✔
88	{
89	regex r_chunk("@@ -([0-9]+),?([0-9]) \\+([0-9]+),?([0-9]) @@.*");	19✔
90
91	if (r_chunk.match(*tok_iter) != 4)	19✔
92	{
93	log("unified_diff") << *tok_iter << r_chunk.size();	2✔
94	return false;	2✔
95	}
96
97	m_range[0] = wex::stoi(r_chunk[0]);	17✔
98	m_range[1] = wex::stoi(r_chunk[1]);	17✔
99	m_range[2] = wex::stoi(r_chunk[2]);	17✔
100	m_range[3] = wex::stoi(r_chunk[3]);	17✔
101
102	m_text.fill({});	17✔
103
104	// Now get all - lines and all + lines, collect them, and invoke callback.
105	CHANGES_LINES(1, 0);	33!
106	CHANGES_LINES(3, 1);	29!
107
108	log::trace("diff parse")	34✔
109	<< std::to_underlying(m_type) << m_path[0].string() << m_diffs;	17✔
110	if (!report_diff())	17!
111	{
112	return false;	×
113	}
114
115	m_is_first = false;	17✔
116
117	if (++tok_iter != tokens.end() && !(*tok_iter).starts_with("@@"))	17!
118	{
119	m_is_last = true;	6✔
120	report_diff();	6✔
121	break; // this was last chunk, continue with header lines	6✔
122	}
123	}	19✔
124	}
125
126	m_is_last = true;	8✔
127
128	if (m_type != diff_t::UNKNOWN)	8✔
129	{
130	m_type = diff_t::LAST;	6✔
131	log::trace("diff parse LAST") << m_path[0].string() << " " << m_diffs;	12✔
132	}
133
134	report_diff();	8✔
135	report_diff_finish();	8✔
136
137	return true;	8✔
138	}	11✔
139
140	bool wex::factory::unified_diff::parse_header(	28✔
141	const std::string& r,
142	const std::string& line,
143	path& p)
144	{
145	regex re(r);	28✔
146
147	if (!re.match(line))	28!
148	{
149	log("unified_diff") << line << re.match_data().text();	×
150	return false;	×
151	}
152
153	p = path(re[0]);	28✔
154
155	return true;	28✔
156	}	28✔

antonvw / wex / 19616694313

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