chugalug 0.0.1

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -0,0 +1 @@
1
+ v0.0.1. First version.
data/LICENSE ADDED
@@ -0,0 +1,184 @@
1
+ Academic Free License (AFL) v. 3.0
2
+
3
+ This Academic Free License (the "License") applies to any original work
4
+ of authorship (the "Original Work") whose owner (the "Licensor") has
5
+ placed the following licensing notice adjacent to the copyright notice
6
+ for the Original Work:
7
+
8
+ Licensed under the Academic Free License version 3.0
9
+
10
+ 1) Grant of Copyright License. Licensor grants You a worldwide,
11
+ royalty-free, non-exclusive, sublicensable license, for the duration of
12
+ the copyright, to do the following:
13
+
14
+ a) to reproduce the Original Work in copies, either alone or as part of
15
+ a collective work;
16
+
17
+ b) to translate, adapt, alter, transform, modify, or arrange the
18
+ Original Work, thereby creating derivative works ("Derivative Works")
19
+ based upon the Original Work;
20
+
21
+ c) to distribute or communicate copies of the Original Work and
22
+ Derivative Works to the public, under any license of your choice that
23
+ does not contradict the terms and conditions, including Licensor's
24
+ reserved rights and remedies, in this Academic Free License;
25
+
26
+ d) to perform the Original Work publicly; and
27
+
28
+ e) to display the Original Work publicly.
29
+
30
+ 2) Grant of Patent License. Licensor grants You a worldwide,
31
+ royalty-free, non-exclusive, sublicensable license, under patent claims
32
+ owned or controlled by the Licensor that are embodied in the Original
33
+ Work as furnished by the Licensor, for the duration of the patents, to
34
+ make, use, sell, offer for sale, have made, and import the Original Work
35
+ and Derivative Works.
36
+
37
+ 3) Grant of Source Code License. The term "Source Code" means the
38
+ preferred form of the Original Work for making modifications to it and
39
+ all available documentation describing how to modify the Original Work.
40
+ Licensor agrees to provide a machine-readable copy of the Source Code of
41
+ the Original Work along with each copy of the Original Work that
42
+ Licensor distributes. Licensor reserves the right to satisfy this
43
+ obligation by placing a machine-readable copy of the Source Code in an
44
+ information repository reasonably calculated to permit inexpensive and
45
+ convenient access by You for as long as Licensor continues to distribute
46
+ the Original Work.
47
+
48
+ 4) Exclusions From License Grant. Neither the names of Licensor, nor the
49
+ names of any contributors to the Original Work, nor any of their
50
+ trademarks or service marks, may be used to endorse or promote products
51
+ derived from this Original Work without express prior permission of the
52
+ Licensor. Except as expressly stated herein, nothing in this License
53
+ grants any license to Licensor's trademarks, copyrights, patents, trade
54
+ secrets or any other intellectual property. No patent license is granted
55
+ to make, use, sell, offer for sale, have made, or import embodiments of
56
+ any patent claims other than the licensed claims defined in Section 2.
57
+ No license is granted to the trademarks of Licensor even if such marks
58
+ are included in the Original Work. Nothing in this License shall be
59
+ interpreted to prohibit Licensor from licensing under terms different
60
+ from this License any Original Work that Licensor otherwise would have a
61
+ right to license.
62
+
63
+ 5) External Deployment. The term "External Deployment" means the use,
64
+ distribution, or communication of the Original Work or Derivative Works
65
+ in any way such that the Original Work or Derivative Works may be used
66
+ by anyone other than You, whether those works are distributed or
67
+ communicated to those persons or made available as an application
68
+ intended for use over a network. As an express condition for the grants
69
+ of license hereunder, You must treat any External Deployment by You of
70
+ the Original Work or a Derivative Work as a distribution under section
71
+ 1(c).
72
+
73
+ 6) Attribution Rights. You must retain, in the Source Code of any
74
+ Derivative Works that You create, all copyright, patent, or trademark
75
+ notices from the Source Code of the Original Work, as well as any
76
+ notices of licensing and any descriptive text identified therein as an
77
+ "Attribution Notice." You must cause the Source Code for any Derivative
78
+ Works that You create to carry a prominent Attribution Notice reasonably
79
+ calculated to inform recipients that You have modified the Original
80
+ Work.
81
+
82
+ 7) Warranty of Provenance and Disclaimer of Warranty. Licensor warrants
83
+ that the copyright in and to the Original Work and the patent rights
84
+ granted herein by Licensor are owned by the Licensor or are sublicensed
85
+ to You under the terms of this License with the permission of the
86
+ contributor(s) of those copyrights and patent rights. Except as
87
+ expressly stated in the immediately preceding sentence, the Original
88
+ Work is provided under this License on an "AS IS" BASIS and WITHOUT
89
+ WARRANTY, either express or implied, including, without limitation, the
90
+ warranties of non-infringement, merchantability or fitness for a
91
+ particular purpose. THE ENTIRE RISK AS TO THE QUALITY OF THE ORIGINAL
92
+ WORK IS WITH YOU. This DISCLAIMER OF WARRANTY constitutes an essential
93
+ part of this License. No license to the Original Work is granted by this
94
+ License except under this disclaimer.
95
+
96
+ 8) Limitation of Liability. Under no circumstances and under no legal
97
+ theory, whether in tort (including negligence), contract, or otherwise,
98
+ shall the Licensor be liable to anyone for any indirect, special,
99
+ incidental, or consequential damages of any character arising as a
100
+ result of this License or the use of the Original Work including,
101
+ without limitation, damages for loss of goodwill, work stoppage,
102
+ computer failure or malfunction, or any and all other commercial damages
103
+ or losses. This limitation of liability shall not apply to the extent
104
+ applicable law prohibits such limitation.
105
+
106
+ 9) Acceptance and Termination. If, at any time, You expressly assented
107
+ to this License, that assent indicates your clear and irrevocable
108
+ acceptance of this License and all of its terms and conditions. If You
109
+ distribute or communicate copies of the Original Work or a Derivative
110
+ Work, You must make a reasonable effort under the circumstances to
111
+ obtain the express assent of recipients to the terms of this License.
112
+ This License conditions your rights to undertake the activities listed
113
+ in Section 1, including your right to create Derivative Works based upon
114
+ the Original Work, and doing so without honoring these terms and
115
+ conditions is prohibited by copyright law and international treaty.
116
+ Nothing in this License is intended to affect copyright exceptions and
117
+ limitations (including "fair use" or "fair dealing"). This License shall
118
+ terminate immediately and You may no longer exercise any of the rights
119
+ granted to You by this License upon your failure to honor the conditions
120
+ in Section 1(c).
121
+
122
+ 10) Termination for Patent Action. This License shall terminate
123
+ automatically and You may no longer exercise any of the rights granted
124
+ to You by this License as of the date You commence an action, including
125
+ a cross-claim or counterclaim, against Licensor or any licensee alleging
126
+ that the Original Work infringes a patent. This termination provision
127
+ shall not apply for an action alleging patent infringement by
128
+ combinations of the Original Work with other software or hardware.
129
+
130
+ 11) Jurisdiction, Venue and Governing Law. Any action or suit relating
131
+ to this License may be brought only in the courts of a jurisdiction
132
+ wherein the Licensor resides or in which Licensor conducts its primary
133
+ business, and under the laws of that jurisdiction excluding its
134
+ conflict-of-law provisions. The application of the United Nations
135
+ Convention on Contracts for the International Sale of Goods is expressly
136
+ excluded. Any use of the Original Work outside the scope of this License
137
+ or after its termination shall be subject to the requirements and
138
+ penalties of copyright or patent law in the appropriate jurisdiction.
139
+ This section shall survive the termination of this License.
140
+
141
+ 12) Attorneys' Fees. In any action to enforce the terms of this License
142
+ or seeking damages relating thereto, the prevailing party shall be
143
+ entitled to recover its costs and expenses, including, without
144
+ limitation, reasonable attorneys' fees and costs incurred in connection
145
+ with such action, including any appeal of such action. This section
146
+ shall survive the termination of this License.
147
+
148
+ 13) Miscellaneous. If any provision of this License is held to be
149
+ unenforceable, such provision shall be reformed only to the extent
150
+ necessary to make it enforceable.
151
+
152
+ 14) Definition of "You" in This License. "You" throughout this License,
153
+ whether in upper or lower case, means an individual or a legal entity
154
+ exercising rights under, and complying with all of the terms of, this
155
+ License. For legal entities, "You" includes any entity that controls, is
156
+ controlled by, or is under common control with you. For purposes of this
157
+ definition, "control" means (i) the power, direct or indirect, to cause
158
+ the direction or management of such entity, whether by contract or
159
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
160
+ outstanding shares, or (iii) beneficial ownership of such entity.
161
+
162
+ 15) Right to Use. You may use the Original Work in all ways not
163
+ otherwise restricted or conditioned by this License or by law, and
164
+ Licensor promises not to interfere with or be responsible for such uses
165
+ by You.
166
+
167
+ 16) Modification of This License. This License is Copyright (c) 2005
168
+ Lawrence Rosen. Permission is granted to copy, distribute, or
169
+ communicate this License without modification. Nothing in this License
170
+ permits You to modify this License as applied to the Original Work or to
171
+ Derivative Works. However, You may modify the text of this License and
172
+ copy, distribute or communicate your modified version (the "Modified
173
+ License") and apply it to other original works of authorship subject to
174
+ the following conditions: (i) You may not indicate in any way that your
175
+ Modified License is the "Academic Free License" or "AFL" and you may not
176
+ use those names in the name of your Modified License; (ii) You must
177
+ replace the notice specified in the first paragraph above with the
178
+ notice "Licensed under <insert your license name here>" or with a notice
179
+ of your own that is not confusingly similar to the notice in this
180
+ License; and (iii) You may not claim that your original works are open
181
+ source software unless your Modified License has been approved by Open
182
+ Source Initiative (OSI) and You comply with its license review and
183
+ certification process.
184
+
@@ -0,0 +1,15 @@
1
+ CHANGELOG
2
+ LICENSE
3
+ Manifest
4
+ README
5
+ Rakefile
6
+ ext/chugalug.c
7
+ ext/csv.h
8
+ ext/extconf.rb
9
+ ext/libcsv.c
10
+ test/data.csv
11
+ test/data_small.csv
12
+ test/easy.csv
13
+ test/messages.csv
14
+ test/quoted_values.csv
15
+ test/unit/test_chugalug.rb
data/README ADDED
@@ -0,0 +1,12 @@
1
+
2
+ Ccsv
3
+
4
+ A pure-C CSV parser.
5
+
6
+ == License
7
+
8
+ Copyright 2010 Josh Ferguson, based off of Ccsv with the License Below:
9
+
10
+ == Attribution
11
+
12
+ Copyright 2007, 2008 Cloudburst, LLC. Licensed under the AFL 3. See the included LICENSE file.
@@ -0,0 +1,9 @@
1
+ require 'echoe'
2
+
3
+ Echoe.new("chugalug") do |p|
4
+ p.author = "Josh Ferguson"
5
+ p.email = "josh@besquared.net"
6
+ p.project = "Besquared"
7
+ p.summary = "A wrapper around libcsv."
8
+ p.url = "http://www.github.com/besquared/chugalug/"
9
+ end
@@ -0,0 +1,32 @@
1
+ # -*- encoding: utf-8 -*-
2
+
3
+ Gem::Specification.new do |s|
4
+ s.name = %q{chugalug}
5
+ s.version = "0.0.1"
6
+
7
+ s.required_rubygems_version = Gem::Requirement.new(">= 1.2") if s.respond_to? :required_rubygems_version=
8
+ s.authors = ["Josh Ferguson"]
9
+ s.date = %q{2010-05-15}
10
+ s.description = %q{A wrapper around libcsv.}
11
+ s.email = %q{josh@besquared.net}
12
+ s.extensions = ["ext/extconf.rb"]
13
+ s.extra_rdoc_files = ["CHANGELOG", "LICENSE", "README", "ext/chugalug.c", "ext/csv.h", "ext/extconf.rb", "ext/libcsv.c"]
14
+ s.files = ["CHANGELOG", "LICENSE", "Manifest", "README", "Rakefile", "ext/chugalug.c", "ext/csv.h", "ext/extconf.rb", "ext/libcsv.c", "test/data.csv", "test/data_small.csv", "test/easy.csv", "test/messages.csv", "test/quoted_values.csv", "test/unit/test_chugalug.rb", "chugalug.gemspec"]
15
+ s.homepage = %q{http://www.github.com/besquared/chugalug/}
16
+ s.rdoc_options = ["--line-numbers", "--inline-source", "--title", "Chugalug", "--main", "README"]
17
+ s.require_paths = ["lib", "ext"]
18
+ s.rubyforge_project = %q{Besquared}
19
+ s.rubygems_version = %q{1.3.6}
20
+ s.summary = %q{A wrapper around libcsv.}
21
+ s.test_files = ["test/unit/test_chugalug.rb"]
22
+
23
+ if s.respond_to? :specification_version then
24
+ current_version = Gem::Specification::CURRENT_SPECIFICATION_VERSION
25
+ s.specification_version = 3
26
+
27
+ if Gem::Version.new(Gem::RubyGemsVersion) >= Gem::Version.new('1.2.0') then
28
+ else
29
+ end
30
+ else
31
+ end
32
+ end
@@ -0,0 +1,79 @@
1
+ #include <ruby.h>
2
+ #include <stdio.h>
3
+ #include <stdlib.h>
4
+ #include <string.h>
5
+ #include <errno.h>
6
+
7
+ #include "csv.h"
8
+
9
+ struct rowdata {
10
+ VALUE ary;
11
+ };
12
+
13
+ void fieldcb(void* str, size_t len, void* data) {
14
+ VALUE ary = ((struct rowdata *)data)->ary;
15
+ rb_ary_store(ary, RARRAY_LEN(ary), rb_str_new(str, len));
16
+ }
17
+
18
+ void rowcb(int c, void* data) {
19
+ rb_yield(((struct rowdata *)data)->ary);
20
+ ((struct rowdata *)data)->ary = rb_ary_new();
21
+ }
22
+
23
+ static int is_space(unsigned char c) {
24
+ if (c == CSV_SPACE || c == CSV_TAB) return 1;
25
+ return 0;
26
+ }
27
+
28
+ static int is_term(unsigned char c) {
29
+ if (c == CSV_CR || c == CSV_LF) return 1;
30
+ return 0;
31
+ }
32
+
33
+ static VALUE foreach(VALUE self, VALUE filename) {
34
+ FILE *file = fopen(StringValueCStr(filename), "r");
35
+
36
+ if (file == NULL) {
37
+ rb_raise(rb_eRuntimeError, "File not found");
38
+ }
39
+
40
+ char buf[1024];
41
+ size_t bytes_read;
42
+ struct csv_parser p;
43
+ unsigned char options = 0;
44
+
45
+ struct rowdata data;
46
+ data.ary = rb_ary_new();
47
+
48
+ if(csv_init(&p, options) != 0) {
49
+ rb_raise(rb_eRuntimeError, "Failed to initialize csv parser");
50
+ exit(EXIT_FAILURE);
51
+ }
52
+
53
+ csv_set_space_func(&p, is_space);
54
+ csv_set_term_func(&p, is_term);
55
+
56
+ while((bytes_read = fread(buf, 1, 1024, file)) > 0) {
57
+ if(csv_parse(&p, buf, bytes_read, fieldcb, rowcb, &data) != bytes_read) {
58
+ rb_raise(rb_eRuntimeError, "Error while parsing file"); //, csv_strerror(csv_error(&p)));
59
+ }
60
+ }
61
+
62
+ csv_fini(&p, fieldcb, rowcb, &data);
63
+ csv_free(&p);
64
+
65
+ if(ferror(file)) {
66
+ fclose(file);
67
+ rb_raise(rb_eRuntimeError, "Error while reading file");
68
+ } else {
69
+ fclose(file);
70
+ }
71
+
72
+ return Qnil;
73
+ }
74
+
75
+ static VALUE rb_cC;
76
+ void Init_chugalug() {
77
+ rb_cC = rb_define_class("Chugalug", rb_cObject);
78
+ rb_define_singleton_method(rb_cC, "foreach", foreach, 1);
79
+ }
@@ -0,0 +1,86 @@
1
+ #ifndef LIBCSV_H__
2
+ #define LIBCSV_H__
3
+ #include <stdlib.h>
4
+ #include <stdio.h>
5
+
6
+ #ifdef __cplusplus
7
+ extern "C" {
8
+ #endif
9
+
10
+ #define CSV_MAJOR 3
11
+ #define CSV_MINOR 0
12
+ #define CSV_RELEASE 0
13
+
14
+ /* Error Codes */
15
+ #define CSV_SUCCESS 0
16
+ #define CSV_EPARSE 1 /* Parse error in strict mode */
17
+ #define CSV_ENOMEM 2 /* Out of memory while increasing buffer size */
18
+ #define CSV_ETOOBIG 3 /* Buffer larger than SIZE_MAX needed */
19
+ #define CSV_EINVALID 4 /* Invalid code,should never be received from csv_error*/
20
+
21
+
22
+ /* parser options */
23
+ #define CSV_STRICT 1 /* enable strict mode */
24
+ #define CSV_REPALL_NL 2 /* report all unquoted carriage returns and linefeeds */
25
+ #define CSV_STRICT_FINI 4 /* causes csv_fini to return CSV_EPARSE if last
26
+ field is quoted and doesn't containg ending
27
+ quote */
28
+ #define CSV_APPEND_NULL 8 /* Ensure that all fields are null-ternimated */
29
+
30
+
31
+ /* Character values */
32
+ #define CSV_TAB 0x09
33
+ #define CSV_SPACE 0x20
34
+ #define CSV_CR 0x0d
35
+ #define CSV_LF 0x0a
36
+ #define CSV_COMMA 0x2c
37
+ #define CSV_QUOTE 0x22
38
+
39
+ struct csv_parser {
40
+ int pstate; /* Parser state */
41
+ int quoted; /* Is the current field a quoted field? */
42
+ size_t spaces; /* Number of continious spaces after quote or in a non-quoted field */
43
+ unsigned char * entry_buf; /* Entry buffer */
44
+ size_t entry_pos; /* Current position in entry_buf (and current size of entry) */
45
+ size_t entry_size; /* Size of entry buffer */
46
+ int status; /* Operation status */
47
+ unsigned char options;
48
+ unsigned char quote_char;
49
+ unsigned char delim_char;
50
+ int (*is_space)(unsigned char);
51
+ int (*is_term)(unsigned char);
52
+ size_t blk_size;
53
+ void *(*malloc_func)(size_t);
54
+ void *(*realloc_func)(void *, size_t);
55
+ void (*free_func)(void *);
56
+ };
57
+
58
+ /* Function Prototypes */
59
+ int csv_init(struct csv_parser *p, unsigned char options);
60
+ int csv_fini(struct csv_parser *p, void (*cb1)(void *, size_t, void *), void (*cb2)(int, void *), void *data);
61
+ void csv_free(struct csv_parser *p);
62
+ int csv_error(struct csv_parser *p);
63
+ char * csv_strerror(int error);
64
+ size_t csv_parse(struct csv_parser *p, const void *s, size_t len, void (*cb1)(void *, size_t, void *), void (*cb2)(int, void *), void *data);
65
+ size_t csv_write(void *dest, size_t dest_size, const void *src, size_t src_size);
66
+ int csv_fwrite(FILE *fp, const void *src, size_t src_size);
67
+ size_t csv_write2(void *dest, size_t dest_size, const void *src, size_t src_size, unsigned char quote);
68
+ int csv_fwrite2(FILE *fp, const void *src, size_t src_size, unsigned char quote);
69
+ int csv_get_opts(struct csv_parser *p);
70
+ int csv_set_opts(struct csv_parser *p, unsigned char options);
71
+ void csv_set_delim(struct csv_parser *p, unsigned char c);
72
+ void csv_set_quote(struct csv_parser *p, unsigned char c);
73
+ unsigned char csv_get_delim(struct csv_parser *p);
74
+ unsigned char csv_get_quote(struct csv_parser *p);
75
+ void csv_set_space_func(struct csv_parser *p, int (*f)(unsigned char));
76
+ void csv_set_term_func(struct csv_parser *p, int (*f)(unsigned char));
77
+ void csv_set_realloc_func(struct csv_parser *p, void *(*)(void *, size_t));
78
+ void csv_set_free_func(struct csv_parser *p, void (*)(void *));
79
+ void csv_set_blk_size(struct csv_parser *p, size_t);
80
+ size_t csv_get_buffer_size(struct csv_parser *p);
81
+
82
+ #ifdef __cplusplus
83
+ }
84
+ #endif
85
+
86
+ #endif
@@ -0,0 +1,3 @@
1
+ require 'mkmf'
2
+ dir_config('chugalug')
3
+ create_makefile('chugalug')
@@ -0,0 +1,579 @@
1
+ /*
2
+ libcsv - parse and write csv data
3
+ Copyright (C) 2008 Robert Gamble
4
+
5
+ This library is free software; you can redistribute it and/or
6
+ modify it under the terms of the GNU Lesser General Public
7
+ License as published by the Free Software Foundation; either
8
+ version 2.1 of the License, or (at your option) any later version.
9
+
10
+ This library is distributed in the hope that it will be useful,
11
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
12
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
13
+ Lesser General Public License for more details.
14
+
15
+ You should have received a copy of the GNU Lesser General Public
16
+ License along with this library; if not, write to the Free Software
17
+ Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
18
+ */
19
+
20
+ #if ___STDC_VERSION__ >= 199901L
21
+ # include <stdint.h>
22
+ #else
23
+ # define SIZE_MAX ((size_t)-1) /* C89 doesn't have stdint.h or SIZE_MAX */
24
+ #endif
25
+
26
+ #include "csv.h"
27
+
28
+ #define VERSION "3.0.0"
29
+
30
+ #define ROW_NOT_BEGUN 0
31
+ #define FIELD_NOT_BEGUN 1
32
+ #define FIELD_BEGUN 2
33
+ #define FIELD_MIGHT_HAVE_ENDED 3
34
+
35
+ /*
36
+ Explanation of states
37
+ ROW_NOT_BEGUN There have not been any fields encountered for this row
38
+ FIELD_NOT_BEGUN There have been fields but we are currently not in one
39
+ FIELD_BEGUN We are in a field
40
+ FIELD_MIGHT_HAVE_ENDED
41
+ We encountered a double quote inside a quoted field, the
42
+ field is either ended or the quote is literal
43
+ */
44
+
45
+ #define MEM_BLK_SIZE 128
46
+
47
+ #define SUBMIT_FIELD(p) \
48
+ do { \
49
+ if (!quoted) \
50
+ entry_pos -= spaces; \
51
+ if (p->options & CSV_APPEND_NULL) \
52
+ ((p)->entry_buf[entry_pos+1]) = '\0'; \
53
+ if (cb1) \
54
+ cb1(p->entry_buf, entry_pos, data); \
55
+ pstate = FIELD_NOT_BEGUN; \
56
+ entry_pos = quoted = spaces = 0; \
57
+ } while (0)
58
+
59
+ #define SUBMIT_ROW(p, c) \
60
+ do { \
61
+ if (cb2) \
62
+ cb2(c, data); \
63
+ pstate = ROW_NOT_BEGUN; \
64
+ entry_pos = quoted = spaces = 0; \
65
+ } while (0)
66
+
67
+ #define SUBMIT_CHAR(p, c) ((p)->entry_buf[entry_pos++] = (c))
68
+
69
+ static char *csv_errors[] = {"success",
70
+ "error parsing data while strict checking enabled",
71
+ "memory exhausted while increasing buffer size",
72
+ "data size too large",
73
+ "invalid status code"};
74
+
75
+ int
76
+ csv_error(struct csv_parser *p)
77
+ {
78
+ /* Return the current status of the parser */
79
+ return p->status;
80
+ }
81
+
82
+ char *
83
+ csv_strerror(int status)
84
+ {
85
+ /* Return a textual description of status */
86
+ if (status >= CSV_EINVALID || status < 0)
87
+ return csv_errors[CSV_EINVALID];
88
+ else
89
+ return csv_errors[status];
90
+ }
91
+
92
+ int
93
+ csv_get_opts(struct csv_parser *p)
94
+ {
95
+ /* Return the currently set options of parser */
96
+ if (p == NULL)
97
+ return -1;
98
+
99
+ return p->options;
100
+ }
101
+
102
+ int
103
+ csv_set_opts(struct csv_parser *p, unsigned char options)
104
+ {
105
+ /* Set the options */
106
+ if (p == NULL)
107
+ return -1;
108
+
109
+ p->options = options;
110
+ return 0;
111
+ }
112
+
113
+ int
114
+ csv_init(struct csv_parser *p, unsigned char options)
115
+ {
116
+ /* Initialize a csv_parser object returns 0 on success, -1 on error */
117
+ if (p == NULL)
118
+ return -1;
119
+
120
+ p->entry_buf = NULL;
121
+ p->pstate = ROW_NOT_BEGUN;
122
+ p->quoted = 0;
123
+ p->spaces = 0;
124
+ p->entry_pos = 0;
125
+ p->entry_size = 0;
126
+ p->status = 0;
127
+ p->options = options;
128
+ p->quote_char = CSV_QUOTE;
129
+ p->delim_char = CSV_COMMA;
130
+ p->is_space = NULL;
131
+ p->is_term = NULL;
132
+ p->blk_size = MEM_BLK_SIZE;
133
+ p->malloc_func = NULL;
134
+ p->realloc_func = realloc;
135
+ p->free_func = free;
136
+
137
+ return 0;
138
+ }
139
+
140
+ void
141
+ csv_free(struct csv_parser *p)
142
+ {
143
+ /* Free the entry_buffer of csv_parser object */
144
+ if (p == NULL)
145
+ return;
146
+
147
+ if (p->entry_buf)
148
+ p->free_func(p->entry_buf);
149
+
150
+ p->entry_buf = NULL;
151
+ p->entry_size = 0;
152
+
153
+ return;
154
+ }
155
+
156
+ int
157
+ csv_fini(struct csv_parser *p, void (*cb1)(void *, size_t, void *), void (*cb2)(int c, void *), void *data)
158
+ {
159
+ /* Finalize parsing. Needed, for example, when file does not end in a newline */
160
+ int quoted = p->quoted;
161
+ int pstate = p->pstate;
162
+ size_t spaces = p->spaces;
163
+ size_t entry_pos = p->entry_pos;
164
+
165
+ if (p == NULL)
166
+ return -1;
167
+
168
+
169
+ if (p->pstate == FIELD_BEGUN && p->quoted && p->options & CSV_STRICT && p->options & CSV_STRICT_FINI) {
170
+ /* Current field is quoted, no end-quote was seen, and CSV_STRICT_FINI is set */
171
+ p->status = CSV_EPARSE;
172
+ return -1;
173
+ }
174
+
175
+ switch (p->pstate) {
176
+ case FIELD_MIGHT_HAVE_ENDED:
177
+ p->entry_pos -= p->spaces + 1; /* get rid of spaces and original quote */
178
+ /* Fall-through */
179
+ case FIELD_NOT_BEGUN:
180
+ case FIELD_BEGUN:
181
+ quoted = p->quoted, pstate = p->pstate;
182
+ spaces = p->spaces, entry_pos = p->entry_pos;
183
+ SUBMIT_FIELD(p);
184
+ SUBMIT_ROW(p, -1);
185
+ case ROW_NOT_BEGUN: /* Already ended properly */
186
+ ;
187
+ }
188
+
189
+ /* Reset parser */
190
+ p->spaces = p->quoted = p->entry_pos = p->status = 0;
191
+ p->pstate = ROW_NOT_BEGUN;
192
+
193
+ return 0;
194
+ }
195
+
196
+ void
197
+ csv_set_delim(struct csv_parser *p, unsigned char c)
198
+ {
199
+ /* Set the delimiter */
200
+ if (p) p->delim_char = c;
201
+ }
202
+
203
+ void
204
+ csv_set_quote(struct csv_parser *p, unsigned char c)
205
+ {
206
+ /* Set the quote character */
207
+ if (p) p->quote_char = c;
208
+ }
209
+
210
+ unsigned char
211
+ csv_get_delim(struct csv_parser *p)
212
+ {
213
+ /* Get the delimiter */
214
+ return p->delim_char;
215
+ }
216
+
217
+ unsigned char
218
+ csv_get_quote(struct csv_parser *p)
219
+ {
220
+ /* Get the quote character */
221
+ return p->quote_char;
222
+ }
223
+
224
+ void
225
+ csv_set_space_func(struct csv_parser *p, int (*f)(unsigned char))
226
+ {
227
+ /* Set the space function */
228
+ if (p) p->is_space = f;
229
+ }
230
+
231
+ void
232
+ csv_set_term_func(struct csv_parser *p, int (*f)(unsigned char))
233
+ {
234
+ /* Set the term function */
235
+ if (p) p->is_term = f;
236
+ }
237
+
238
+ void
239
+ csv_set_realloc_func(struct csv_parser *p, void *(*f)(void *, size_t))
240
+ {
241
+ /* Set the realloc function used to increase buffer size */
242
+ if (p && f) p->realloc_func = f;
243
+ }
244
+
245
+ void
246
+ csv_set_free_func(struct csv_parser *p, void (*f)(void *))
247
+ {
248
+ /* Set the free function used to free the buffer */
249
+ if (p && f) p->free_func = f;
250
+ }
251
+
252
+ void
253
+ csv_set_blk_size(struct csv_parser *p, size_t size)
254
+ {
255
+ /* Set the block size used to increment buffer size */
256
+ if (p) p->blk_size = size;
257
+ }
258
+
259
+ size_t
260
+ csv_get_buffer_size(struct csv_parser *p)
261
+ {
262
+ /* Get the size of the entry buffer */
263
+ if (p)
264
+ return p->entry_size;
265
+ return 0;
266
+ }
267
+
268
+ static int
269
+ csv_increase_buffer(struct csv_parser *p)
270
+ {
271
+ /* Increase the size of the entry buffer. Attempt to increase size by
272
+ * p->blk_size, if this is larger than SIZE_MAX try to increase current
273
+ * buffer size to SIZE_MAX. If allocation fails, try to allocate halve
274
+ * the size and try again until successful or increment size is zero.
275
+ */
276
+
277
+ size_t to_add = p->blk_size;
278
+ void *vp;
279
+
280
+ if ( p->entry_size >= SIZE_MAX - to_add )
281
+ to_add = SIZE_MAX - p->entry_size;
282
+
283
+ if (!to_add) {
284
+ p->status = CSV_ETOOBIG;
285
+ return -1;
286
+ }
287
+
288
+ while ((vp = p->realloc_func(p->entry_buf, p->entry_size + to_add)) == NULL) {
289
+ to_add /= 2;
290
+ if (!to_add) {
291
+ p->status = CSV_ENOMEM;
292
+ return -1;
293
+ }
294
+ }
295
+
296
+ /* Update entry buffer pointer and entry_size if successful */
297
+ p->entry_buf = vp;
298
+ p->entry_size += to_add;
299
+ return 0;
300
+ }
301
+
302
+ size_t
303
+ csv_parse(struct csv_parser *p, const void *s, size_t len, void (*cb1)(void *, size_t, void *), void (*cb2)(int c, void *), void *data)
304
+ {
305
+ unsigned const char *us = s; /* Access input data as array of unsigned char */
306
+ unsigned char c; /* The character we are currently processing */
307
+ size_t pos = 0; /* The number of characters we have processed in this call */
308
+
309
+ /* Store key fields into local variables for performance */
310
+ unsigned char delim = p->delim_char;
311
+ unsigned char quote = p->quote_char;
312
+ int (*is_space)(unsigned char) = p->is_space;
313
+ int (*is_term)(unsigned char) = p->is_term;
314
+ int quoted = p->quoted;
315
+ int pstate = p->pstate;
316
+ size_t spaces = p->spaces;
317
+ size_t entry_pos = p->entry_pos;
318
+
319
+
320
+ if (!p->entry_buf && pos < len) {
321
+ /* Buffer hasn't been allocated yet and len > 0 */
322
+ if (csv_increase_buffer(p) != 0) {
323
+ p->quoted = quoted, p->pstate = pstate, p->spaces = spaces, p->entry_pos = entry_pos;
324
+ return pos;
325
+ }
326
+ }
327
+
328
+ while (pos < len) {
329
+ /* Check memory usage, increase buffer if neccessary */
330
+ if (entry_pos == ((p->options & CSV_APPEND_NULL) ? p->entry_size - 1 : p->entry_size) ) {
331
+ if (csv_increase_buffer(p) != 0) {
332
+ p->quoted = quoted, p->pstate = pstate, p->spaces = spaces, p->entry_pos = entry_pos;
333
+ return pos;
334
+ }
335
+ }
336
+
337
+ c = us[pos++];
338
+
339
+ switch (pstate) {
340
+ case ROW_NOT_BEGUN:
341
+ case FIELD_NOT_BEGUN:
342
+ if (is_space ? is_space(c) : c == CSV_SPACE || c == CSV_TAB) { /* Space or Tab */
343
+ continue;
344
+ } else if (is_term ? is_term(c) : c == CSV_CR || c == CSV_LF) { /* Carriage Return or Line Feed */
345
+ if (pstate == FIELD_NOT_BEGUN) {
346
+ SUBMIT_FIELD(p);
347
+ SUBMIT_ROW(p, (unsigned char)c);
348
+ } else { /* ROW_NOT_BEGUN */
349
+ /* Don't submit empty rows by default */
350
+ if (p->options & CSV_REPALL_NL) {
351
+ SUBMIT_ROW(p, (unsigned char)c);
352
+ }
353
+ }
354
+ continue;
355
+ } else if (c == delim) { /* Comma */
356
+ SUBMIT_FIELD(p);
357
+ break;
358
+ } else if (c == quote) { /* Quote */
359
+ pstate = FIELD_BEGUN;
360
+ quoted = 1;
361
+ } else { /* Anything else */
362
+ pstate = FIELD_BEGUN;
363
+ quoted = 0;
364
+ SUBMIT_CHAR(p, c);
365
+ }
366
+ break;
367
+ case FIELD_BEGUN:
368
+ if (c == quote) { /* Quote */
369
+ if (quoted) {
370
+ SUBMIT_CHAR(p, c);
371
+ pstate = FIELD_MIGHT_HAVE_ENDED;
372
+ } else {
373
+ /* STRICT ERROR - double quote inside non-quoted field */
374
+ if (p->options & CSV_STRICT) {
375
+ p->status = CSV_EPARSE;
376
+ p->quoted = quoted, p->pstate = pstate, p->spaces = spaces, p->entry_pos = entry_pos;
377
+ return pos-1;
378
+ }
379
+ SUBMIT_CHAR(p, c);
380
+ spaces = 0;
381
+ }
382
+ } else if (c == delim) { /* Comma */
383
+ if (quoted) {
384
+ SUBMIT_CHAR(p, c);
385
+ } else {
386
+ SUBMIT_FIELD(p);
387
+ }
388
+ } else if (is_term ? is_term(c) : c == CSV_CR || c == CSV_LF) { /* Carriage Return or Line Feed */
389
+ if (!quoted) {
390
+ SUBMIT_FIELD(p);
391
+ SUBMIT_ROW(p, (unsigned char)c);
392
+ } else {
393
+ SUBMIT_CHAR(p, c);
394
+ }
395
+ } else if (!quoted && (is_space? is_space(c) : c == CSV_SPACE || c == CSV_TAB)) { /* Tab or space for non-quoted field */
396
+ SUBMIT_CHAR(p, c);
397
+ spaces++;
398
+ } else { /* Anything else */
399
+ SUBMIT_CHAR(p, c);
400
+ spaces = 0;
401
+ }
402
+ break;
403
+ case FIELD_MIGHT_HAVE_ENDED:
404
+ /* This only happens when a quote character is encountered in a quoted field */
405
+ if (c == delim) { /* Comma */
406
+ entry_pos -= spaces + 1; /* get rid of spaces and original quote */
407
+ SUBMIT_FIELD(p);
408
+ } else if (is_term ? is_term(c) : c == CSV_CR || c == CSV_LF) { /* Carriage Return or Line Feed */
409
+ entry_pos -= spaces + 1; /* get rid of spaces and original quote */
410
+ SUBMIT_FIELD(p);
411
+ SUBMIT_ROW(p, (unsigned char)c);
412
+ } else if (is_space ? is_space(c) : c == CSV_SPACE || c == CSV_TAB) { /* Space or Tab */
413
+ SUBMIT_CHAR(p, c);
414
+ spaces++;
415
+ } else if (c == quote) { /* Quote */
416
+ if (spaces) {
417
+ /* STRICT ERROR - unescaped double quote */
418
+ if (p->options & CSV_STRICT) {
419
+ p->status = CSV_EPARSE;
420
+ p->quoted = quoted, p->pstate = pstate, p->spaces = spaces, p->entry_pos = entry_pos;
421
+ return pos-1;
422
+ }
423
+ spaces = 0;
424
+ SUBMIT_CHAR(p, c);
425
+ } else {
426
+ /* Two quotes in a row */
427
+ pstate = FIELD_BEGUN;
428
+ }
429
+ } else { /* Anything else */
430
+ /* STRICT ERROR - unescaped double quote */
431
+ if (p->options & CSV_STRICT) {
432
+ p->status = CSV_EPARSE;
433
+ p->quoted = quoted, p->pstate = pstate, p->spaces = spaces, p->entry_pos = entry_pos;
434
+ return pos-1;
435
+ }
436
+ pstate = FIELD_BEGUN;
437
+ spaces = 0;
438
+ SUBMIT_CHAR(p, c);
439
+ }
440
+ break;
441
+ default:
442
+ break;
443
+ }
444
+ }
445
+ p->quoted = quoted, p->pstate = pstate, p->spaces = spaces, p->entry_pos = entry_pos;
446
+ return pos;
447
+ }
448
+
449
+ size_t
450
+ csv_write (void *dest, size_t dest_size, const void *src, size_t src_size)
451
+ {
452
+ unsigned char *cdest = dest;
453
+ const unsigned char *csrc = src;
454
+ size_t chars = 0;
455
+
456
+ if (src == NULL)
457
+ return 0;
458
+
459
+ if (cdest == NULL)
460
+ dest_size = 0;
461
+
462
+ if (dest_size > 0)
463
+ *cdest++ = '"';
464
+ chars++;
465
+
466
+ while (src_size) {
467
+ if (*csrc == '"') {
468
+ if (dest_size > chars)
469
+ *cdest++ = '"';
470
+ if (chars < SIZE_MAX) chars++;
471
+ }
472
+ if (dest_size > chars)
473
+ *cdest++ = *csrc;
474
+ if (chars < SIZE_MAX) chars++;
475
+ src_size--;
476
+ csrc++;
477
+ }
478
+
479
+ if (dest_size > chars)
480
+ *cdest = '"';
481
+ if (chars < SIZE_MAX) chars++;
482
+
483
+ return chars;
484
+ }
485
+
486
+ int
487
+ csv_fwrite (FILE *fp, const void *src, size_t src_size)
488
+ {
489
+ const unsigned char *csrc = src;
490
+
491
+ if (fp == NULL || src == NULL)
492
+ return 0;
493
+
494
+ if (fputc('"', fp) == EOF)
495
+ return EOF;
496
+
497
+ while (src_size) {
498
+ if (*csrc == '"') {
499
+ if (fputc('"', fp) == EOF)
500
+ return EOF;
501
+ }
502
+ if (fputc(*csrc, fp) == EOF)
503
+ return EOF;
504
+ src_size--;
505
+ csrc++;
506
+ }
507
+
508
+ if (fputc('"', fp) == EOF) {
509
+ return EOF;
510
+ }
511
+
512
+ return 0;
513
+ }
514
+
515
+ size_t
516
+ csv_write2 (void *dest, size_t dest_size, const void *src, size_t src_size, unsigned char quote)
517
+ {
518
+ unsigned char *cdest = dest;
519
+ const unsigned char *csrc = src;
520
+ size_t chars = 0;
521
+
522
+ if (src == NULL)
523
+ return 0;
524
+
525
+ if (dest == NULL)
526
+ dest_size = 0;
527
+
528
+ if (dest_size > 0)
529
+ *cdest++ = quote;
530
+ chars++;
531
+
532
+ while (src_size) {
533
+ if (*csrc == quote) {
534
+ if (dest_size > chars)
535
+ *cdest++ = quote;
536
+ if (chars < SIZE_MAX) chars++;
537
+ }
538
+ if (dest_size > chars)
539
+ *cdest++ = *csrc;
540
+ if (chars < SIZE_MAX) chars++;
541
+ src_size--;
542
+ csrc++;
543
+ }
544
+
545
+ if (dest_size > chars)
546
+ *cdest = quote;
547
+ if (chars < SIZE_MAX) chars++;
548
+
549
+ return chars;
550
+ }
551
+
552
+ int
553
+ csv_fwrite2 (FILE *fp, const void *src, size_t src_size, unsigned char quote)
554
+ {
555
+ const unsigned char *csrc = src;
556
+
557
+ if (fp == NULL || src == NULL)
558
+ return 0;
559
+
560
+ if (fputc(quote, fp) == EOF)
561
+ return EOF;
562
+
563
+ while (src_size) {
564
+ if (*csrc == quote) {
565
+ if (fputc(quote, fp) == EOF)
566
+ return EOF;
567
+ }
568
+ if (fputc(*csrc, fp) == EOF)
569
+ return EOF;
570
+ src_size--;
571
+ csrc++;
572
+ }
573
+
574
+ if (fputc(quote, fp) == EOF) {
575
+ return EOF;
576
+ }
577
+
578
+ return 0;
579
+ }