1 # mdcode: extract C code from a _markdown_ file.
3 _markdown_ is a popular format for simple text markup which can easily
4 be converted to HTML. As it allows easy indication of sections of
5 code, it is quite suitable for use in literate programming. This file
6 is an example of that usage.
8 The code included below provides two related functionalities.
9 Firstly it provides a library routine for extracting code out of a
10 _markdown_ file, so that other routines might make use of it.
12 Secondly it provides a simple client of this routine which extracts
13 1 or more C-language files from a markdown document so they can be
14 passed to a C compiler. These two combined to make a tool that is needed
15 to compile this tool. Yes, this is circular. A prototype tool was
16 used for the first extraction.
18 The tool provided is described as specific to the C language as it
21 ##### Example: a _line_ command
23 #line __line-number__ __file-name__
25 lines so that the C compiler will report where in the markdown file
26 any error is found. This tool is suitable for any other language
27 which allows the same directive, or will treat it as a comment.
31 Literate programming is more than just including comments with the
32 code, even nicely formatted comments. It also involves presenting the
33 code in an order that makes sense to a human, rather than an order
34 that makes sense to a compiler. For this reason a core part of any
35 literate programming tool is the ability to re-arrange the code found
36 in the document into a different order in the final code file - or
37 files. This requires some form of linkage to be encoded.
39 The approach taken here is focused around section headings - of any
42 All the code in any section is treated as a single sequential
43 collection of code, and is named by the section that it is in. If
44 multiple sections have the same name, then the code blocks in all of
45 them are joined together in the order they appear in the document.
47 A code section can contain a special marker which starts with 2
49 The text after the marker must be the name of some section which
50 contains code. Code from that section will be interpolated in place
51 of the marker, and will be indented to match the indent of the marker.
53 It is not permitted for the same code to be interpolated multiple
54 times. Allowing this might make some sense, but it is probably a
55 mistake, and prohibiting it make some of the code a bit cleaner.
57 Equally, every section of code should be interpolated at least once -
58 with two exceptions. These exceptions are imposed by the tool, not
59 the library. A different client could impose different rules on the
60 names of top-level code sections.
62 The first exception we have already seen. A section name starting
63 __Example:__ indicates code that is not to be included in the final product.
65 The second exception is for the top level code sections which will be
66 written to files. Again these are identified by their section name.
67 This must start with __File:__ the following text (after optional
68 spaces) will be used as a file name.
70 Any section containing code that does not start __Example:__ or
71 __File:__ must be included in some other section exactly once.
75 Allowing multiple top level code sections which name different files
76 means that one _markdown_ document can describe several files. This
77 is very useful with the C language where a program file and a header
78 file might be related. For the present document we will have a header
79 file and two code files, one with the library content and one for the
82 It will also be very convenient to create a `makefile` fragment to
83 ensure the code is compiled correctly. A simple `make -f mdcode.mk`
84 will "do the right thing".
90 mdcode.h libmdcode.c md2c.c mdcode.mk : mdcode.mdc
108 ## internal functions
113 libmdcode.o : libmdcode.c mdcode.h
114 $(CC) $(CFLAGS) -c libmdcode.c
130 md2c : md2c.o libmdcode.o
131 $(CC) $(CFLAGS) -o md2c md2c.o libmdcode.o
132 md2c.o : md2c.c mdcode.h
133 $(CC) $(CFLAGS) -c md2c.c
137 As the core purpose of _mdcode_ is to discover and re-arrange blocks
138 of text, it makes sense to map the whole document file into memory and
139 produce a data structure which lists various parts of the file in the
140 appropriate order. Each node in this structure will have some text
141 from the document, a child pointer, and a next pointer, any of which
142 might not be present. The text is most easily stored as a pointer and a
143 length. We'll call this a `text`
145 A list of these `code_nodes` will belong to each section and it will
146 be useful to have a separate `section` data structure to store the
147 list of `code_nodes`, the section name, and some other information.
149 This other information will include a reference counter so we can
150 ensure proper referencing, and an `indent` depth. As referenced
151 content can have an extra indent added, we need to know what that is.
152 The `code_node` will also have an `indent` depth which eventually gets
153 set to the sum for the indents from all references on the path from
165 struct code_node *code;
166 struct section *next;
173 struct code_node *next;
174 struct section *child;
181 struct code_node *last;
186 You will note that the `struct psection` contains an anonymous `struct
187 section` embedded at the start. To make this work right, GCC
188 requires the `-fplan9-extensions` flag.
190 ##### File: mdcode.mk
192 CFLAGS += -fplan9-extensions
194 ### Manipulating the node
196 Though a tree with `next` and `child` links is the easiest way to
197 assemble the various code sections, it is not the easiest form for
198 using them. For that a simple list would be best.
200 So once we have a fully linked File section we will want to linearize
201 it, so that the `child` links become `NULL` and the `next` links will
202 find everything required. It is at this stage that the requirements
203 that each section is linked only once becomes import.
205 `code_linearize` will merge the `code_node`s from any child into the
206 given `code_node`. As it does this it sets the 'indent' field for
209 Note that we don't clear the section's `last` pointer, even though
210 it no longer owns any code. This allows subsequent code to see if a
211 section ever had any code, and to report an error if a section is
212 referenced but not defined.
214 ##### internal functions
216 static void code_linearize(struct code_node *code)
219 for (t = code; t; t = t->next)
221 for (; code; code = code->next)
223 struct code_node *next = code->next;
224 struct psection *pchild =
225 (struct psection *)code->child;
226 int indent = pchild->indent;
227 code->next = code->child->code;
228 code->child->code = NULL;
230 for (t = code; t->next; t = t->next)
231 t->next->indent = code->indent + indent;
236 Once a client has made use of a linearized code set, it will probably
239 void code_free(struct code_node *code)
242 struct code_node *this;
244 code_linearize(code);
251 ##### exported functions
253 void code_free(struct code_node *code);
255 ### Building the tree
257 As we parse the document there are two things we will want to do to
258 node trees: add some text or add a reference. We'll assume for now
259 that the relevant section structures have been found, and will just
260 deal with the `code_node`.
262 Adding text simply means adding another node. We will never have
263 empty nodes, even if the last node only has a child, new text must go
266 ##### internal functions
268 static void code_add_text(struct psection *where, struct text txt,
274 n = malloc(sizeof(*n));
277 n->line_no = line_no;
281 where->last->next = n;
287 However when adding a link, we might be able to include it in the last
288 `code_node` if it currently only has text.
290 void code_add_link(struct psection *where, struct psection *to,
296 to->refcnt++; // this will be checked elsewhere
297 if (where->last && where->last->child == NULL) {
298 where->last->child = to;
301 n = malloc(sizeof(*n));
308 where->last->next = n;
316 Now we need a lookup table to be able to find sections by name.
317 Something that provides an `n*log(N)` search time is probably
318 justified, but for now I want a minimal stand-alone program so a
319 linked list managed by insertion-sort will do. As a comparison
320 function it is easiest to sort based on length before content. So
321 sections won't be in standard lexical order, but that isn't important.
323 If we cannot find a section, we simply want to create it. This allows
324 sections and references to be created in any order. Sections with
325 no references or no content will cause a warning eventually.
327 #### internal functions
329 static int text_cmp(struct text a, struct text b)
332 return a.len - b.len;
333 return strncmp(a.txt, b.txt, a.len);
336 static struct psection *section_find(struct psection **list, struct text name)
338 struct psection *new;
340 int cmp = text_cmp((*list)->section, name);
345 list = (struct psection **)&((*list)->next);
347 /* Add this section */
348 new = malloc(sizeof(*new));
359 ## Parsing the _markdown_
361 Parsing markdown is fairly easy, though there are complications.
363 The document is divided into "paragraphs" which are mostly separated by blank
364 lines (which may contain white space). The first few characters of
365 the first line of a paragraph determine the type of paragraph. For
366 our purposes we are only interested in list paragraphs, code
367 paragraphs, section headings, and everything else. Section headings
368 are single-line paragraphs and so do not require a preceding or
369 following blank line.
371 Section headings start with 1 or more hash characters (__#__). List
372 paragraphs start with hyphen, asterisk, plus, or digits followed by a
373 period. Code paragraphs aren't quite so easy.
375 The "standard" code paragraph starts with 4 or more spaces, or a tab.
376 However if the previous paragraph was a list paragraph, then those
377 spaces indicate another paragraph in the same list item, and 8 or
378 more spaces are required. Unless a nested list is in effect, in
379 which case 12 or more are need. Unfortunately not all _markdown_
380 parsers agree on nested lists.
382 Two alternate styles for marking code are in active use. "Github" uses
383 three backticks(_`` ``` ``_), while "pandoc" uses three or more tildes
384 (_~~~_). In these cases the code should not be indented.
386 Trying to please everyone as much as possible, this parser will handle
387 everything except for code inside lists.
389 So an indented (4+) paragraph after a list paragraph is always a list
390 paragraph, otherwise it is a code paragraph. A paragraph that starts
391 with three backticks or three tildes is code which continues until a
392 matching string of backticks or tildes.
396 While walking the document looking for various markers we will *not*
397 use the `struct text` introduced earlier as advancing that requires
398 updating both start and length which feels clumsy. Instead we will
399 carry `pos` and `end` pointers, only the first of which needs to
402 So to start, we need to skip various parts of the document. `lws`
403 stands for "Linear White Space" and is a term that comes from the
404 Email RFCs (e.g. RFC822). `line` and `para` are self explanatory.
405 Note that `skip_para` needs to update the current line number.
406 `skip_line` doesn't but every caller should.
408 #### internal functions
410 static char *skip_lws(char *pos, char *end)
412 while (pos < end && (*pos == ' ' || *pos == '\t'))
417 static char *skip_line(char *pos, char *end)
419 while (pos < end && *pos != '\n')
426 static char *skip_para(char *pos, char *end, int *line_no)
428 /* Might return a pointer to a blank line, as only
429 * one trailing blank line is skipped
432 pos = skip_line(pos, end);
438 *(pos = skip_lws(pos, end)) != '\n') {
439 pos = skip_line(pos, end);
442 if (pos < end && *pos == '\n') {
449 ### Recognising things
451 Recognising a section header is trivial and doesn't require a
452 function. However we need to extract the content of a section header
453 as a `struct text` for passing to `section_find`.
454 Recognising the start of a new list is fairly easy. Recognising the
455 start (and end) of code is a little messy so we provide a function for
456 matching the first few characters, which has a special case for "4
459 #### internal includes
464 #### internal functions
466 static struct text take_header(char *pos, char *end)
470 while (pos < end && *pos == '#')
472 while (pos < end && *pos == ' ')
475 while (pos < end && *pos != '\n')
477 while (pos > section.txt &&
478 (pos[-1] == '#' || pos[-1] == ' '))
480 section.len = pos - section.txt;
484 static int is_list(char *pos, char *end)
486 if (strchr("-*+", *pos))
489 while (pos < end && isdigit(*pos))
491 if (pos < end && *pos == '.')
497 static int matches(char *start, char *pos, char *end)
500 return matches("\t", pos, end) ||
501 matches(" ", pos, end);
502 return (pos + strlen(start) < end &&
503 strncmp(pos, start, strlen(start)) == 0);
506 ### Extracting the code
508 Now that we can skip paragraphs and recognise what type each paragraph
509 is, it is time to parse the file and extract the code. We'll do this
510 in two parts, first we look at what to do with some code once we
511 find it, and then how to actually find it.
513 When we have some code, we know where it is, what the end marker
514 should look like, and which section it is in.
516 There are two sorts of end markers: the presence of a particular
517 string, or the absence of an indent. We will use a string to
518 represent a presence, and a `NULL` to represent the absence.
520 While looking at code we don't think about paragraphs are all - just
521 look for a line that starts with the right thing.
522 Every line that is still code then needs to be examined to see if it
523 is a section reference.
525 When a section reference is found, all preceding code (if any) must be
526 added to the current section, then the reference is added.
528 When we do find the end of the code, all text that we have found but
529 not processed needs to be saved too.
531 When adding a reference we need to set the `indent`. This is the
532 number of spaces (counting 8 for tabs) after the natural indent of the
533 code (which is a tab or 4 spaces). We use a separate function `count_spaces`
536 #### internal functions
538 static int count_space(char *sol, char *p)
552 static char *take_code(char *pos, char *end, char *marker,
553 struct psection **table, struct text section,
557 int line_no = *line_nop;
558 int start_line = line_no;
559 struct psection *sect;
561 sect = section_find(table, section);
567 if (marker && matches(marker, pos, end))
570 (skip_lws(pos, end))[0] != '\n' &&
571 !matches(NULL, pos, end))
572 /* Paragraph not indented */
575 /* Still in code - check for reference */
580 else if (strcmp(sol, " ") == 0)
583 t = skip_lws(sol, end);
584 if (t[0] != '#' || t[1] != '#') {
585 /* Just regular code here */
586 pos = skip_line(sol, end);
594 txt.len = pos - start;
595 code_add_text(sect, txt, start_line);
597 ref = take_header(t, end);
599 struct psection *refsec = section_find(table, ref);
600 code_add_link(sect, refsec, count_space(sol, t));
602 pos = skip_line(t, end);
605 start_line = line_no;
610 txt.len = pos - start;
611 code_add_text(sect, txt, start_line);
614 pos = skip_line(pos, end);
623 It is when looking for the code that we actually use the paragraph
624 structure. We need to recognise section headings so we can record the
625 name, list paragraphs so we can ignore indented follow-on paragraphs,
626 and the three different markings for code.
628 #### internal functions
630 static struct psection *code_find(char *pos, char *end)
632 struct psection *table = NULL;
635 struct text section = {0};
639 section = take_header(pos, end);
641 pos = skip_line(pos, end);
643 } else if (is_list(pos, end)) {
645 pos = skip_para(pos, end, &line_no);
646 } else if (!in_list && matches(NULL, pos, end)) {
647 pos = take_code(pos, end, NULL, &table,
649 } else if (matches("```", pos, end)) {
651 pos = skip_line(pos, end);
653 pos = take_code(pos, end, "```", &table,
655 } else if (matches("~~~", pos, end)) {
657 pos = skip_line(pos, end);
659 pos = take_code(pos, end, "~~~", &table,
664 pos = skip_para(pos, end, &line_no);
670 ### Returning the code
672 Having found all the code blocks and gathered them into a list of
673 section, we are now ready to return them to the caller. This is where
674 to perform consistency checks, like at most one reference and at least
675 one definition for each section.
677 All the sections with no references are returned in a list for the
678 caller to consider. The are linearized first so that the substructure
679 is completely hidden -- except for the small amount of structure
680 displayed in the line numbers.
682 To return errors, we have the caller pass a function which takes an
683 error message - a `code_err_fn`.
687 typedef void (*code_err_fn)(char *msg);
689 #### internal functions
690 struct section *code_extract(char *pos, char *end, code_err_fn error)
692 struct psection *table;
693 struct section *result = NULL;
694 struct section *tofree = NULL;
696 table = code_find(pos, end);
699 struct psection *t = (struct psection*)table->next;
700 if (table->last == NULL) {
703 "Section \"%.*s\" is referenced but not declared",
704 table->section.len, table->section.txt);
708 if (table->refcnt == 0) {
709 /* Root-section, return it */
710 table->next = result;
712 code_linearize(result->code);
714 table->next = tofree;
716 if (table->refcnt > 1) {
719 "Section \"%.*s\" referenced multiple times (%d).",
720 table->section.len, table->section.txt,
729 struct section *t = tofree->next;
736 ##### exported functions
738 struct section *code_extract(char *pos, char *end, code_err_fn error);
743 Now that we can extract code from a document and link it all together
744 it is time to do something with that code. Firstly we need to print
747 ### Printing the Code
749 Printing is mostly straight forward - we just walk the list and print
750 the code sections, adding whatever indent is required for each line.
751 However there is a complication (isn't there always)?
753 For code that was recognised because the paragraph was indented, we
754 need to strip that indent first. For other code, we don't.
756 The approach taken here is simple, though it could arguably be wrong
757 in some unlikely cases. So it might need to be fixed later.
759 If the first line of a code block is indented, then either one tab or
760 4 spaces are striped from every non-blank line.
762 This could go wrong if the first line of a code block marked by
763 _`` ``` ``_ is indented. To overcome this we would need to
764 record someextra state in each `code_node`. For now we won't bother.
766 The indents we insert will all be spaces. This might not work well
769 ##### client functions
771 static void code_print(FILE *out, struct code_node *node,
774 for (; node; node = node->next) {
775 char *c = node->code.txt;
776 int len = node->code.len;
782 fprintf(out, "#line %d \"%s\"\n",
783 node->line_no, fname);
784 if (*c == ' ' || *c == '\t')
787 fprintf(out, "%*s", node->indent, "");
789 if (*c == '\t' && len > 1) {
792 } else if (strncmp(c, " ", 4) == 0 && len > 4) {
801 } while (len && c[-1] != '\n');
806 ### Bringing it all together
808 We are just about ready for the `main` function of the tool which will
809 extract all this lovely code and compile it. Just one helper is still
812 #### Handling filenames
814 Section names are stored in `struct text` which is not `nul`
815 terminated. Filenames passed to `open` need to be null terminated.
816 So we need to convert one to the other, and strip the leading `File:`
817 of while we are at it.
819 ##### client functions
821 static void copy_fname(char *name, int space, struct text t)
826 if (len < 5 || strncmp(sec, "File:", 5) != 0)
830 while (len && sec[0] == ' ') {
836 strncpy(name, sec, len);
842 And now we take a single file name, extract the code, and if there are
843 no error we write out a file for each appropriate code section. And
847 ##### client includes
851 #include <sys/mman.h>
855 ##### client functions
858 static void pr_err(char *msg)
861 fprintf(stderr, "%s\n", msg);
864 int main(int argc, char *argv[])
869 struct section *table, *s, *prev;
873 fprintf(stderr, "Usage: mdcode file.mdc\n");
876 fd = open(argv[1], O_RDONLY);
878 fprintf(stderr, "mdcode: cannot open %s: %s\n",
879 argv[1], strerror(errno));
882 len = lseek(fd, 0, 2);
883 file = mmap(NULL, len, PROT_READ, MAP_SHARED, fd, 0);
884 table = code_extract(file, file+len, pr_err);
887 (code_free(s->code), prev = s, s = s->next, free(prev))) {
890 if (strncmp(s->section.txt, "Example:", 8) == 0)
892 if (strncmp(s->section.txt, "File:", 5) != 0) {
893 fprintf(stderr, "Unreferenced section is not a file name: %.*s\n",
894 s->section.len, s->section.txt);
898 copy_fname(fname, sizeof(fname), s->section);
900 fprintf(stderr, "Missing file name at:%.*s\n",
901 s->section.len, s->section.txt);
905 fl = fopen(fname, "w");
907 fprintf(stderr, "Cannot create %s: %s\n",
908 fname, strerror(errno));
912 code_print(fl, s->code, argv[1]);