/++
module source_read_source_files;
- open markup files
- if master file scan for addional files to import/insert
+/
module doc_reform.io_in.read_source_files;
template spineRawMarkupContent() {
import
std.file,
std.path,
doc_reform.meta,
doc_reform.io_in.paths_source,
doc_reform.meta.rgx;
mixin spineRgxIn;
static auto rgx = RgxI();
string[] _images=[];
@safe string[] _extract_images(S)(S content_block) {
string[] images_;
string _content_block = content_block.to!string;
if (auto m = _content_block.matchAll(rgx.image)) {
images_ ~= m.captures[1].to!string;
}
return images_;
}
auto rawsrc = RawMarkupContent();
alias ContentsInsertsImages = Tuple!(
char[][], "contents",
string[], "insert_files",
string[], "images"
);
alias HeaderContentInsertsImages = Tuple!(
char[], "header",
char[][], "src_txt",
string[], "insert_files",
string[], "images"
);
@safe auto spineRawMarkupContent(O,Fn)(O _opt_action, Fn fn_src) {
auto _0_header_1_body_content_2_insert_filelist_tuple
= rawsrc.sourceContentSplitIntoHeaderAndBody(_opt_action, rawsrc.sourceContent(fn_src), fn_src);
return _0_header_1_body_content_2_insert_filelist_tuple;
}
struct RawMarkupContent {
@safe final sourceContent(in string fn_src) {
auto raw = MarkupRawUnit();
string source_txt_str
= raw.markupSourceReadIn(fn_src);
return source_txt_str;
}
@safe final auto sourceContentSplitIntoHeaderAndBody(O)(
O _opt_action,
in string source_txt_str,
in string fn_src=""
) {
auto raw = MarkupRawUnit();
string[] insert_file_list;
string[] images_list;
HeaderContentInsertsImages t
= raw.markupSourceHeaderContentRawLineTupleArray(source_txt_str);
char[] header_raw = t.header;
char[][] sourcefile_body_content = t.src_txt;
if (fn_src.match(rgx.src_fn_master)) { // filename with path needed if master file (.ssm) not otherwise
auto ins = Inserts();
ContentsInsertsImages tu
= ins.scan_master_src_for_insert_files_and_import_content(_opt_action, sourcefile_body_content, fn_src);
sourcefile_body_content = tu.contents;
insert_file_list = tu.insert_files.dup;
images_list = tu.images.dup;
} else if (_opt_action.source || _opt_action.pod) {
auto ins = Inserts();
ContentsInsertsImages tu
= ins.scan_master_src_for_insert_files_and_import_content(_opt_action, sourcefile_body_content, fn_src);
images_list = tu.images.dup;
}
string header_type = "";
t = tuple(
header_raw,
sourcefile_body_content,
insert_file_list,
images_list
);
return t;
}
}
struct MarkupRawUnit {
import std.file;
@safe final private string readInMarkupSource(in char[] fn_src) {
enforce(
exists(fn_src) != 0,
"file not found: «" ~
fn_src ~ "»"
);
string source_txt_str;
try {
if (exists(fn_src)) {
if (fn_src.getLinkAttributes.attrIsFile) {
source_txt_str = fn_src.readText;
} else {
}
}
} catch (ErrnoException ex) {
} catch (UTFException ex) {
// Handle validation errors
} catch (FileException ex) {
// Handle errors
}
std.utf.validate(source_txt_str);
return source_txt_str;
}
@trusted final private char[][] header0Content1(in string src_text) { // cast(char[])
/+ split string on _first_ match of "^:?A~\s" into [header, content] array/tuple +/
char[][] header_and_content;
auto m = (cast(char[]) src_text).matchFirst(rgx.heading_a);
header_and_content ~= m.pre;
header_and_content ~= m.hit ~ m.post;
assert(header_and_content.length == 2,
"document markup is broken, header body split == "
~ header_and_content.length.to!string
~ "; (header / body array split should == 2 (split is on level A~))"
);
return header_and_content;
}
@trusted final private char[][] markupSourceLineArray(in char[] src_text) { // cast(char[])
char[][] source_line_arr
= (cast(char[]) src_text).split(rgx.newline_eol_strip_preceding);
return source_line_arr;
}
@safe string markupSourceReadIn(in string fn_src) {
static auto rgx = RgxI();
enforce(
fn_src.match(rgx.src_pth_sst_or_ssm),
"not a dr markup filename: «" ~
fn_src ~ "»"
);
string source_txt_str = readInMarkupSource(fn_src);
return source_txt_str;
}
@safe HeaderContentInsertsImages markupSourceHeaderContentRawLineTupleArray(in string source_txt_str) {
string[] file_insert_list = [];
string[] images_list = [];
char[][] hc = header0Content1(source_txt_str);
char[] header = hc[0];
char[] source_txt = hc[1];
char[][] source_line_arr = markupSourceLineArray(source_txt);
HeaderContentInsertsImages t = tuple(
header,
source_line_arr,
file_insert_list,
images_list
);
return t;
}
@safe final char[][] getInsertMarkupSourceContentRawLineArray(
in char[] fn_src_insert,
Regex!(char) rgx_file
) {
enforce(
fn_src_insert.match(rgx_file),
"not a dr markup filename: «" ~
fn_src_insert ~ "»"
);
string source_txt_str = readInMarkupSource(fn_src_insert);
char[][] source_line_arr = markupSourceLineArray(source_txt_str);
return source_line_arr;
}
}
struct Inserts {
alias ContentsAndImages = Tuple!(
char[][], "insert_contents",
string[], "images"
);
@safe ContentsAndImages scan_subdoc_source(O)(
O _opt_action,
char[][] markup_sourcefile_insert_content,
string fn_src
) {
mixin spineRgxDocStructFlags;
char[][] contents_insert;
int code_block_status = 0;
enum codeBlock { off, curly, tic, }
int[string] type = flags_type_init;
auto fn_pth_full = fn_src.match(rgx.src_pth_sst_or_ssm);
auto markup_src_file_path = fn_pth_full.captures[1];
foreach (line; markup_sourcefile_insert_content) {
if (code_block_status == codeBlock.curly) {
type["header_make"] = 0;
type["header_meta"] = 0;
if (line.matchFirst(rgx.block_curly_code_close)) {
code_block_status = codeBlock.off;
}
contents_insert ~= line;
} else if (line.matchFirst(rgx.block_curly_code_open)) {
code_block_status = codeBlock.curly;
type["header_make"] = 0;
type["header_meta"] = 0;
contents_insert ~= line;
} else if (code_block_status == codeBlock.tic) {
type["header_make"] = 0;
type["header_meta"] = 0;
if (line.matchFirst(rgx.block_tic_close)) {
code_block_status = codeBlock.off;
}
contents_insert ~= line;
} else if (line.matchFirst(rgx.block_tic_code_open)) {
code_block_status = codeBlock.tic;
type["header_make"] = 0;
type["header_meta"] = 0;
contents_insert ~= line;
} else if (auto m = line.match(rgx.insert_src_fn_ssi_or_sst)) {
type["header_make"] = 0;
type["header_meta"] = 0;
auto insert_fn = m.captures[2];
auto insert_sub_pth = m.captures[1];
auto fn_src_insert
= chainPath(markup_src_file_path, insert_sub_pth ~ insert_fn).array;
auto raw = MarkupRawUnit();
auto markup_sourcesubfile_insert_content
= raw.getInsertMarkupSourceContentRawLineArray(fn_src_insert, rgx.src_fn_find_inserts);
debug(insert_file) {
writeln(line);
writeln(fn_src_insert);
writeln(
" length contents insert array: ",
markup_sourcesubfile_insert_content.length
);
}
if (_opt_action.source || _opt_action.pod) {
_images ~= _extract_images(markup_sourcesubfile_insert_content);
}
auto ins = Inserts();
/+
- 1. load file
- 2. read lines
- 3. scan lines
- a. if filename insert, and insert filename
- repeat 1
- b. else
- add line to new array;
- build image list, search for any image files to add to image list
+/
} else {
type["header_make"] = 0;
type["header_meta"] = 0;
contents_insert ~= line; // images to extract for image list?
if (_opt_action.source || _opt_action.pod) {
string[] _image_linelist = _extract_images(line);
if (_image_linelist.length > 0) {
_images ~= _image_linelist;
}
}
}
} // end src subdoc (inserts) loop
ContentsAndImages t = tuple(
contents_insert,
_images
);
return t;
}
@safe ContentsInsertsImages scan_master_src_for_insert_files_and_import_content(O)(
O _opt_action,
char[][] sourcefile_body_content,
string fn_src
) {
import std.algorithm;
mixin spineRgxDocStructFlags;
char[][] contents;
int code_block_status = 0;
enum codeBlock { off, curly, tic, }
auto fn_pth_full = fn_src.match(rgx.src_pth_sst_or_ssm);
auto markup_src_file_path = fn_pth_full.captures[1];
char[][] contents_insert;
string[] _images =[];
string[] insert_file_list =[];
foreach (line; sourcefile_body_content) {
if (code_block_status == codeBlock.curly) {
if (line.matchFirst(rgx.block_curly_code_close)) {
code_block_status = codeBlock.off;
}
contents ~= line;
} else if (line.matchFirst(rgx.block_curly_code_open)) {
code_block_status = codeBlock.curly;
contents ~= line;
} else if (code_block_status == codeBlock.tic) {
if (line.matchFirst(rgx.block_tic_close)) {
code_block_status = codeBlock.off;
}
contents ~= line;
} else if (line.matchFirst(rgx.block_tic_code_open)) {
code_block_status = codeBlock.tic;
contents ~= line;
} else if (auto m = line.match(rgx.insert_src_fn_ssi_or_sst)) {
auto insert_fn = m.captures[2];
auto insert_sub_pth = m.captures[1];
auto fn_src_insert
= chainPath(markup_src_file_path, insert_sub_pth ~ insert_fn).array;
insert_file_list ~= fn_src_insert.to!string;
auto raw = MarkupRawUnit();
/+ TODO +/
auto markup_sourcefile_insert_content
= raw.getInsertMarkupSourceContentRawLineArray(fn_src_insert, rgx.src_fn_find_inserts);
debug(insert_file) {
writeln(line);
writeln(fn_src_insert);
writeln(
" length contents insert array: ",
markup_sourcefile_insert_content.length
);
}
auto ins = Inserts();
ContentsAndImages contents_insert_tu = ins.scan_subdoc_source(
_opt_action,
markup_sourcefile_insert_content,
fn_src_insert.to!string
);
contents ~= contents_insert_tu.insert_contents;
if (_opt_action.source || _opt_action.pod) {
string[] _image_linelist = _extract_images(contents_insert_tu.images);
if (_image_linelist.length > 0) {
_images ~= _image_linelist;
}
}
/+
- 1. load file
- 2. read lines
- 3. scan lines
- a. if filename insert, and insert filename
- repeat 1
- b. else
- add line to new array;
- build image list, search for any image files to add to image list
+/
} else {
contents ~= line;
if (_opt_action.source || _opt_action.pod) {
string[] _image_linelist = _extract_images(line);
if (_image_linelist.length > 0) {
_images ~= _image_linelist;
}
}
}
} // end src doc loop
string[] images = [];
foreach(i; uniq(_images.sort())) {
images ~= i;
}
debug(insert_file) {
writeln(__LINE__);
writeln(contents.length);
}
ContentsInsertsImages t = tuple(
contents,
insert_file_list,
images
);
return t;
}
}
}