-*- mode: org -*- #+TITLE: sisudoc spine (doc_reform) output sqlite #+DESCRIPTION: documents - structuring, publishing in multiple formats & search #+FILETAGS: :spine:output:db:sql:sqlite: #+AUTHOR: Ralph Amissah #+EMAIL: [[mailto:ralph.amissah@gmail.com][ralph.amissah@gmail.com]] #+COPYRIGHT: Copyright (C) 2015 (continuously updated, current 2026) Ralph Amissah #+LANGUAGE: en #+STARTUP: content hideblocks hidestars noindent entitiespretty #+PROPERTY: header-args+ :eval never-export :exports code #+PROPERTY: header-args+ :noweb yes :padline no #+PROPERTY: header-args+ :results silent :cache no #+PROPERTY: header-args+ :mkdirp yes #+OPTIONS: H:3 num:nil toc:t \n:t ::t |:t ^:nil -:t f:t *:t - magic single double-quote → " ← FIX changes hilighting behavior (occuring after it) in org document. INVESTIGATE (org-mode CONFIG?) FIND & FIX - [[./doc-reform.org][doc-reform.org]] [[./][org/]] - [[./output_hub.org][output_hub]] sudo chown ralph:ralph /var/www - create db ~dr/bin/spine-ldc -v \ --sqlite-db-create --sqlite-db-filename="spine.search.db" \ --output=/var/www/html \ ~grotto/repo/git.repo/code/project-spine/doc-reform-markup/markup_samples/markup/pod-samples/pod/* - update db ~dr/bin/spine-ldc -v \ --sqlite-update --sqlite-db-filename="spine.search.db" \ --output=/var/www/html \ ~grotto/repo/git.repo/code/project-spine/doc-reform-markup/markup_samples/markup/pod-samples/pod/* - produce html (and some other) output files ~dr/bin/spine-ldc -v --html --epub --latex --odt \ --output=/var/www \ ~grotto/repo/git.repo/code/project-spine/doc-reform-markup/markup_samples/markup/pod-samples/pod/* * sql ** _module, templates_ :module: *** template #+HEADER: :tangle "../src/sisudoc/outputs/io_out/sqlite_collection_db.d" #+HEADER: :noweb yes #+BEGIN_SRC d <> module sisudoc.outputs.io_out.sqlite_collection_db; <> <> <> <> <> <> <> <> <> <> <> <> <> <> #+END_SRC *** hub **** common #+NAME: sqlite_common #+BEGIN_SRC d mixin spineRgxOut; mixin spineRgxXHTML; mixin InternalMarkup; static auto rgx = RgxO(); static auto rgx_xhtml = RgxXHTML(); static auto mkup = InlineMarkup(); long _metadata_tid_lastrowid; #+END_SRC **** sqlite schema #+NAME: sqlite_schema #+HEADER: :noweb yes #+BEGIN_SRC d /+ ↓ sqlite db schema version, stamped into the db as PRAGMA user_version bump whenever the table or index definitions in this module change, and bump the matching expectation in sisudoc-spine-search-cgi +/ enum sqlite_db_schema_version = <>; /+ ↓ tally of failed sqlite statements, shared: sqlite output may run in parallel +/ private shared int _sqlite_failure_tally = 0; /+ ↓ schema version mismatch is reported once per run, not once per document +/ private shared int _sqlite_schema_mismatch_reported = 0; /+ ↓ record a failed sqlite statement, reported on stderr, tallied for exit status +/ void sqliteFailureRecord(string note, string msg) { import core.atomic : atomicOp; atomicOp!"+="(_sqlite_failure_tally, 1); stderr.writeln("ERROR SQLite (", note, "): ", msg); } /+ ↓ count of failed sqlite statements this run, zero when all went well +/ int sqliteFailureTally() { import core.atomic : atomicLoad; return atomicLoad(_sqlite_failure_tally); } /+ ↓ compare the schema version stamped in an existing db with this spine's +/ void sqliteSchemaVersionCheck(Db)(Db db, string sqlite_file) { import core.atomic : atomicOp; int _found; try { _found = db.execute("PRAGMA user_version;").oneValue!int; } catch (Exception) { return; } if (_found == sqlite_db_schema_version || _found == 0) { /+ ↓ zero: db predates the stamp, or has no tables yet +/ return; } if (atomicOp!"+="(_sqlite_schema_mismatch_reported, 1) == 1) { stderr.writeln("ERROR SQLite : db schema version mismatch: ", sqlite_file); stderr.writeln(" db holds schema version ", _found, ", this spine writes schema version ", sqlite_db_schema_version); stderr.writeln(" recreate the db, e.g.: spine --sqlite-db-recreate", " --sqlite-db-path=... --sqlite-db-filename=..."); sqliteFailureRecord("schema version", "db schema version " ~ _found.to!string ~ " != spine schema version " ~ sqlite_db_schema_version.to!string); } } #+END_SRC **** collection #+NAME: template_sqlite_collection #+HEADER: :noweb yes #+BEGIN_SRC d template SQLiteHubBuildTablesAndPopulate() { void SQLiteHubBuildTablesAndPopulate(D)(D doc) { auto pth_sqlite = spinePathsSQLite!()(doc.matters.sqlite.filename, doc.matters.sqlite.path); if ((isValidPath(pth_sqlite.base) && exists(pth_sqlite.base) != 0 && pth_sqlite.base.isDir)) { } else { try { pth_sqlite.base.mkdirRecurse; } catch (FileException ex) { stderr.writeln("FATAL: cannot create --sqlite-db-path directory: ", pth_sqlite.base); stderr.writeln(" (", ex.msg, ")"); import core.runtime; core.runtime.Runtime.terminate(); } } template SQLiteDbStatementComposite() { void SQLiteDbStatementComposite(Db,D)( Db db, D doc ) { <> if (doc.matters.opt.action.vox_gt_1) { writeln(" ", pth_sqlite.sqlite_file); } } } try { auto db = Database(pth_sqlite.sqlite_file); sqliteSchemaVersionCheck(db, pth_sqlite.sqlite_file); SQLiteDbStatementComposite!()(db, doc); } catch (FileException e) { writeln("Failed (FileException): ", e.msg, " ", pth_sqlite.sqlite_file); writeln(e.file, " line: ", e.line); import core.runtime; core.runtime.Runtime.terminate(); } catch (ErrnoException e) { writeln("Failed (ErrnoException): ", e.msg, " ", pth_sqlite.sqlite_file); writeln(e.file, " line: ", e.line); import core.runtime; core.runtime.Runtime.terminate(); } catch (Exception e) { writeln("Failed (Exception): ", e.msg, " ", pth_sqlite.sqlite_file); writeln(e.file, " line: ", e.line); import core.runtime; core.runtime.Runtime.terminate(); } catch (Throwable) { writeln("Failed (Trowable): ", pth_sqlite.sqlite_file); import core.runtime; core.runtime.Runtime.terminate(); } } } #+END_SRC **** discrete #+NAME: template_sqlite_discrete #+HEADER: :noweb yes #+BEGIN_SRC d template SQLiteHubDiscreteBuildTablesAndPopulate() { void SQLiteHubDiscreteBuildTablesAndPopulate(D)(D doc) { auto url_html = spineUrlsHTML!()(doc.matters.conf_make_meta.conf.w_srv_data_root_url_html, doc.matters.src.language); auto pth_sqlite = spinePathsSQLiteDiscrete!()(doc.matters.output_path, doc.matters.src.language); // doc.matters.db_path if ((isValidPath(pth_sqlite.base) && exists(pth_sqlite.base) != 0 && pth_sqlite.base.isDir)) { } else { try { pth_sqlite.base.mkdirRecurse; } catch (FileException ex) { stderr.writeln("FATAL: cannot create --sqlite-db-path directory: ", pth_sqlite.base); stderr.writeln(" (", ex.msg, ")"); import core.runtime; core.runtime.Runtime.terminate(); } } auto db = Database(pth_sqlite.sqlite_file(doc.matters.src.filename)); sqliteSchemaVersionCheck(db, pth_sqlite.sqlite_file(doc.matters.src.filename)); template SQLiteDiscreteDbStatementComposite() { void SQLiteDiscreteDbStatementComposite(Db,D)( Db db, D doc ) { try { <> } catch (FileException e) { writeln("Failed (FileException): ", e.msg); writeln(e.file, " line: ", e.line); import core.runtime; core.runtime.Runtime.terminate(); } catch (ErrnoException e) { writeln("Failed (ErrnoException): ", e.msg); writeln(e.file, " line: ", e.line); import core.runtime; core.runtime.Runtime.terminate(); } catch (Exception e) { writeln("Failed (Exception): ", e.msg); writeln(e.file, " line: ", e.line); import core.runtime; core.runtime.Runtime.terminate(); } catch (Throwable) { import core.runtime; core.runtime.Runtime.terminate(); } if (doc.matters.opt.action.vox_gt_1) { writeln(" ", pth_sqlite.sqlite_file(doc.matters.src.filename)); } } } SQLiteDiscreteDbStatementComposite!()(db, doc); } } #+END_SRC *** db run #+NAME: template_sqlite_run #+BEGIN_SRC d template SQLiteDbRun() { void SQLiteDbRun(Db,St,O)( Db db, St db_statement, O opt_action, string note, ) { debug(sql_statement) { writeln(db_statement); } try { db.run( "\nBEGIN TRANSACTION;\n" ~ db_statement ~ "\nCOMMIT TRANSACTION;\n" ); } catch (ErrnoException ex) { sqliteFailureRecord(note, ex.msg); } catch (Exception ex) { sqliteFailureRecord(note, ex.msg); } { /+ debug +/ if (opt_action.debug_do_sqlite) { writeln(note); if (opt_action.vox_gt_3) { writeln(db_statement); } } } } } #+END_SRC *** munge #+NAME: templates_sqlite_munge #+HEADER: :noweb yes #+BEGIN_SRC d template SQLinsertDelimiter() { string SQLinsertDelimiter(string _txt) { _txt = _txt .replaceAll(rgx.quotation_mark_sql_insert_delimiter, "$0$0"); return _txt; } } template SQLiteFormatAndLoadObject() { auto SQLiteFormatAndLoadObject(M)(M doc_matters) { mixin spineRgxOut; mixin spineRgxXHTML; struct sqlite_format_and_load_objects { <> <> <> <> <> <> <> <> <> <> <> <> <> <> <> <> <> <> <> <> <> <> <> <> <> <> <> <> } return sqlite_format_and_load_objects(); } } #+END_SRC *** sqlite instructions **** create #+NAME: template_sqlite_create #+HEADER: :noweb yes #+BEGIN_SRC d template SQLiteTablesReCreate() { string SQLiteTablesReCreate()() { string _sql_instruct; _sql_instruct = format(q"┃ <> <> <> <> <> <> PRAGMA user_version = %s; ┃", sqlite_db_schema_version, ); return _sql_instruct; } } #+END_SRC **** delete #+NAME: template_sqlite_delete #+HEADER: :noweb yes #+BEGIN_SRC d template SQLiteDeleteDocument() { string SQLiteDeleteDocument(M)(M doc_matters) { <> <> <> return _delete_uid; } } #+END_SRC **** insert metadata #+NAME: template_sqlite_insert_metadata #+HEADER: :noweb yes #+BEGIN_SRC d template SQLiteInsertMetadata() { string SQLiteInsertMetadata(M)(M doc_matters) { <> <> <> <> return _insert_metadata; } } #+END_SRC **** insert metadata topics #+NAME: template_sqlite_insert_metadata_topics #+HEADER: :noweb yes #+BEGIN_SRC d template SQLiteInsertMetadataTopics() { string SQLiteInsertMetadataTopics(M)(M doc_matters) { <> <> <> <> <> } return _insert_topics.join.to!(char[]).toUTF8; } } #+END_SRC **** insert doc objects loop #+NAME: template_sqlite_insert_doc_objects_loop #+HEADER: :noweb yes #+BEGIN_SRC d template SQLiteInsertDocObjectsLoop() { string SQLiteInsertDocObjectsLoop(D)(D doc) { string _uid = SQLinsertDelimiter!()(doc.matters.src.doc_uid); auto url_html = spineUrlsHTML!()(doc.matters.conf_make_meta.conf.w_srv_data_root_url_html, doc.matters.src.language); string insertDocObjectsRow(O)(O obj) { <> <> <> <> return _insert_doc_objects_row; } <> } } #+END_SRC **** tables create #+NAME: template_sqlite_tables_create #+HEADER: :noweb yes #+BEGIN_SRC d template SQLiteTablesCreate() { void SQLiteTablesCreate(E,O,C)(E env, O opt_action, C config) { import d2sqlite3; <> try { <> } catch (FileException e) { writeln("Failed (FileException): ", e.msg); writeln(e.file, " line: ", e.line); import core.runtime; core.runtime.Runtime.terminate(); } catch (ErrnoException e) { writeln("Failed (ErrnoException): ", e.msg); writeln(e.file, " line: ", e.line); import core.runtime; core.runtime.Runtime.terminate(); } catch (Exception e) { writeln("Failed (Exception): ", e.msg); writeln(e.file, " line: ", e.line); import core.runtime; core.runtime.Runtime.terminate(); } catch (Throwable) { import core.runtime; core.runtime.Runtime.terminate(); } } } #+END_SRC **** db create #+NAME: sqlite_db_create #+HEADER: :noweb yes #+BEGIN_SRC d if (opt_action.sqlite_db_create) { string _db_statement; string db_filename = (opt_action.sqliteDB_filename.length > 0) ? opt_action.sqliteDB_filename : (config.conf.w_srv_db_sqlite_filename.length > 0) ? config.conf.w_srv_db_sqlite_filename : ""; string db_path = (opt_action.sqliteDB_path.length > 0) ? opt_action.sqliteDB_path : (config.conf.w_srv_db_sqlite_path.length > 0) ? config.conf.w_srv_db_sqlite_path : ""; if (db_filename.length > 0 && db_path.length > 0) { if (opt_action.vox_gt_3) { writeln("db name: ", db_filename); writeln("db path: ", db_path); writeln("db name & path: ", db_path, "/", db_filename); } if (opt_action.vox_gt_2) { writeln("attempting to create db: ", db_path, "/", db_filename); } auto pth_sqlite = spinePathsSQLite!()(db_filename, db_path); if ((isValidPath(pth_sqlite.base) && exists(pth_sqlite.base) != 0 && pth_sqlite.base.isDir)) { } else { try { pth_sqlite.base.mkdirRecurse; } catch (FileException ex) { stderr.writeln("FATAL: cannot create --sqlite-db-path directory: ", pth_sqlite.base); stderr.writeln(" (", ex.msg, ")"); import core.runtime; core.runtime.Runtime.terminate(); } } auto db = Database(pth_sqlite.sqlite_file); { _db_statement ~= SQLiteTablesReCreate!()(); } SQLiteDbRun!()(db, _db_statement, opt_action, "TABLE RE-CREATE"); } else { writeln("must provide db name & output root path either on the command line or in configuration file"); writeln("db name: ", db_filename); writeln("db path: ", db_path); } } #+END_SRC **** tables drop #+NAME: template_sqlite_tables_drop #+BEGIN_SRC d template SQLiteDbDrop() { void SQLiteDbDrop(O,C)(O opt_action, C config) { writeln("db drop"); if ((opt_action.sqlite_db_drop)) { string db_filename = (opt_action.sqliteDB_filename.length > 0) ? opt_action.sqliteDB_filename : (config.conf.w_srv_db_sqlite_filename.length > 0) ? config.conf.w_srv_db_sqlite_filename : ""; string db_path = (opt_action.sqliteDB_path.length > 0) // ? opt_action.sqliteDB_path : (config.conf.w_srv_db_sqlite_path.length > 0) ? config.conf.w_srv_db_sqlite_path : ""; if (db_filename.length > 0 && db_path.length > 0) { auto pth_sqlite = spinePathsSQLite!()(db_filename, db_path); writeln("remove(", pth_sqlite.sqlite_file, ")"); try { remove(pth_sqlite.sqlite_file); } catch (FileException ex) { // handle error } } else { writeln("must provide db name & output root path either on the command line or in configuration file"); writeln("db name: ", db_filename); writeln("db path: ", db_path); } } } } #+END_SRC ** 1. [#A] sqlite_db_statement :statement: *** collection #+NAME: sqlite_db_statement_composite_collection #+BEGIN_SRC d string _db_statement; if ((doc.matters.opt.action.sqlite_db_create)) { auto pth_sqlite = spinePathsSQLite!()(doc.matters.sqlite.filename, doc.matters.sqlite.path); if ((isValidPath(pth_sqlite.base) && exists(pth_sqlite.base) != 0 && pth_sqlite.base.isDir)) { } else { try { pth_sqlite.base.mkdirRecurse; } catch (FileException ex) { stderr.writeln("FATAL: cannot create --sqlite-db-path directory: ", pth_sqlite.base); stderr.writeln(" (", ex.msg, ")"); import core.runtime; core.runtime.Runtime.terminate(); } } _db_statement ~= SQLiteTablesReCreate!()(); SQLiteDbRun!()(db, _db_statement, doc.matters.opt.action, "TABLE RE-CREATE"); _db_statement = []; } if (doc.matters.opt.action.sqlite_delete) { _db_statement ~= SQLiteDeleteDocument!()(doc.matters); SQLiteDbRun!()(db, _db_statement, doc.matters.opt.action, "DELETE Document"); _db_statement = []; } if (doc.matters.opt.action.sqlite_update) { _db_statement ~= SQLiteDeleteDocument!()(doc.matters); SQLiteDbRun!()(db, _db_statement, doc.matters.opt.action, "DELETE Document"); _db_statement = []; _db_statement ~= SQLiteInsertMetadata!()(doc.matters); SQLiteDbRun!()(db, _db_statement, doc.matters.opt.action, "INSERT MetaData"); _db_statement = []; /+ get tid (lastrowid or max) for use in doc.objects table +/ _db_statement ~= doc.SQLiteInsertDocObjectsLoop!(); SQLiteDbRun!()(db, _db_statement, doc.matters.opt.action, "INSERT DocObjects"); _db_statement = []; _db_statement ~= SQLiteInsertMetadataTopics!()(doc.matters); SQLiteDbRun!()(db, _db_statement, doc.matters.opt.action, "INSERT MetaDataTopics"); _db_statement = []; } db.close; #+END_SRC *** discrete #+NAME: sqlite_db_statement_composite_discrete #+BEGIN_SRC d { string _db_statement; _db_statement ~= SQLiteTablesReCreate!()(); _db_statement ~= SQLiteInsertMetadata!()(doc.matters); _db_statement ~= SQLiteInsertMetadataTopics!()(doc.matters); _db_statement ~= doc.SQLiteInsertDocObjectsLoop!(); SQLiteDbRun!()(db, _db_statement, doc.matters.opt.action, "table CREATE Tables, INSERT DocObjects"); } db.close; #+END_SRC ** 2. imports #+NAME: sqlite_imports #+BEGIN_SRC d import sisudoc.outputs.io_out; import sisudoc.outputs.io_out.rgx; import sisudoc.outputs.io_out.rgx_xhtml; import std.file; import std.uri; // import std.digest.sha; import std.conv : to; import std.typecons : Nullable; import d2sqlite3; #+END_SRC ** 3. format and load template *** 1. prepare objects (munge, sanitize, markup) **** 1. _text_ generic munge (sanitize text for search) - [3/4] (search text, applies to all but code blocks) - [ ] remove whitespace, paragraph on single line (formatting kept so far) - [X] remove font face attributes - [X] move embedded endnotes - [X] place after text object - [X] remove embedded endnote numbers (rely on html output to represent) - [X] urls - [X] clean url markers, leave plain link text - [X] place urls after text object and its endnotes #+NAME: sanitize_text_for_search #+BEGIN_SRC d string generic_munge_sanitize_text_for_search( string _txt, ) { string _notes; string _urls; if (_txt.matchFirst(rgx.inline_notes_al_gen)) { foreach (m; _txt.matchAll(rgx.inline_notes_al_gen_text)) { _notes ~= "\n" ~ m["text"]; } _txt = _txt.replaceAll(rgx.inline_notes_al_gen, ""); } if (_txt.matchFirst(rgx.inline_link)) { foreach (m; _txt.matchAll(rgx.inline_link)) { if (m["link"].match(rgx.url)) { _urls ~= "\n" ~ m["link"]; } } _txt = _txt.replaceAll(rgx.inline_link_clean, ""); } if (_notes.length > 0) { _txt ~= _notes; } if (_urls.length > 0) { _txt ~= _urls; } { /+ debug +/ if (doc_matters.opt.action.debug_do_sqlite) { writeln(_txt, "\n"); } } debug(sql_text_clean) { writeln(_txt); } return _txt; } #+END_SRC **** 2. _html_ light html objects - place endnotes after text object - light inline html markup ***** munge ****** general munge (special characters, inline markup, move notes) #+NAME: sanitize_and_munge_inline_html_munge #+BEGIN_SRC d string munge_html(M,O)( M doc_matters, const O obj, ) { string _html_special_characters(string _txt){ _txt = _txt .replaceAll(rgx_xhtml.ampersand, "&") .replaceAll(rgx_xhtml.quotation, """) .replaceAll(rgx_xhtml.less_than, "<") .replaceAll(rgx_xhtml.greater_than, ">") .replaceAll(rgx.nbsp_char, " ") .replaceAll(rgx.br_line_inline, "
") .replaceAll(rgx.br_line, "
") .replaceAll(rgx.br_line_spaced, "

") .replaceAll(rgx_xhtml.line_break, "
"); return _txt; } string _html_font_face(string _txt){ _txt = _txt .replaceAll(rgx.inline_emphasis, "$1") .replaceAll(rgx.inline_bold, "$1") .replaceAll(rgx.inline_underscore, "$1") .replaceAll(rgx.inline_italics, "$1") .replaceAll(rgx.inline_superscript, "$1") .replaceAll(rgx.inline_subscript, "$1") .replaceAll(rgx.inline_strike, "$1") .replaceAll(rgx.inline_insert, "$1") .replaceAll(rgx.inline_mono, "$1") .replaceAll(rgx.inline_cite, "$1"); return _txt; } string _notes; string _urls; string _txt = _html_font_face(_html_special_characters(obj.text)); { /+ debug +/ if (doc_matters.opt.action.debug_do_sqlite) { writeln(_txt, "\n"); } } return _txt; } #+END_SRC ****** special characters #+NAME: sanitize_and_munge_inline_html_special_characters #+BEGIN_SRC d string html_special_characters(string _txt){ _txt = _txt .replaceAll(rgx_xhtml.ampersand, "&") .replaceAll(rgx_xhtml.quotation, """) .replaceAll(rgx_xhtml.less_than, "<") .replaceAll(rgx_xhtml.greater_than, ">") .replaceAll(rgx.nbsp_char, " ") .replaceAll(rgx.br_line_inline, "
") .replaceAll(rgx.br_line, "
") .replaceAll(rgx.br_line_spaced, "

") .replaceAll(rgx_xhtml.line_break, "
"); return _txt; } #+END_SRC ****** special characters for code #+NAME: sanitize_and_munge_inline_html_special_characters_code #+BEGIN_SRC d string html_special_characters_code(string _txt){ _txt = _txt .replaceAll(rgx_xhtml.ampersand, "&") .replaceAll(rgx_xhtml.quotation, """) .replaceAll(rgx_xhtml.less_than, "<") .replaceAll(rgx_xhtml.greater_than, ">") .replaceAll(rgx.nbsp_char, " "); return _txt; } #+END_SRC ****** font_face #+NAME: sanitize_and_munge_inline_html_font_face #+BEGIN_SRC d string html_font_face(string _txt){ _txt = _txt .replaceAll(rgx.inline_emphasis, "$1") .replaceAll(rgx.inline_bold, "$1") .replaceAll(rgx.inline_underscore, "$1") .replaceAll(rgx.inline_italics, "$1") .replaceAll(rgx.inline_superscript, "$1") .replaceAll(rgx.inline_subscript, "$1") .replaceAll(rgx.inline_strike, "$1") .replaceAll(rgx.inline_insert, "$1") .replaceAll(rgx.inline_mono, "$1") .replaceAll(rgx.inline_cite, "$1"); return _txt; } #+END_SRC ****** inline markup ******* grouped text #+NAME: sanitize_and_munge_inline_html_grouped_text_bullets_indents #+BEGIN_SRC d string inline_grouped_text_bullets_indents(M,O)( M doc_matters, const O obj, string _txt, string _suffix = ".html", string _xml_type = "seg", ) { static auto rgx = RgxO(); static auto rgx_xhtml = RgxXHTML(); if (obj.metainfo.is_a == "group") { _txt = (_txt) .replaceAll(rgx.grouped_para_indent_1, "  ") .replaceAll(rgx.grouped_para_indent_2, "    ") .replaceAll(rgx.grouped_para_indent_3, "      ") .replaceAll(rgx.grouped_para_indent_4, "        ") .replaceAll(rgx.grouped_para_indent_5, "          ") .replaceAll(rgx.grouped_para_indent_6, "            ") .replaceAll(rgx.grouped_para_indent_7, "              ") .replaceAll(rgx.grouped_para_indent_8, "                ") .replaceAll(rgx.grouped_para_indent_9, "                  ") .replaceAll(rgx.grouped_para_indent_hang, "  ") .replaceAll(rgx.grouped_para_bullet, "●  ") .replaceAll(rgx.grouped_para_bullet_indent_1, "  ●  ") .replaceAll(rgx.grouped_para_bullet_indent_2, "    ●  ") .replaceAll(rgx.grouped_para_bullet_indent_3, "      ●  ") .replaceAll(rgx.grouped_para_bullet_indent_4, "        ●  ") .replaceAll(rgx.grouped_para_bullet_indent_5, "          ●  ") .replaceAll(rgx.grouped_para_bullet_indent_6, "            ●  ") .replaceAll(rgx.grouped_para_bullet_indent_7, "              ●  ") .replaceAll(rgx.grouped_para_bullet_indent_8, "                ●  ") .replaceAll(rgx.grouped_para_bullet_indent_9, "                  ●  "); } return _txt; } #+END_SRC ******* images #+NAME: sanitize_and_munge_inline_html_images #+BEGIN_SRC d string inline_images(M,O)( M doc_matters, const O obj, string _txt, string _suffix = ".html", string _xml_type = "seg", ) { string _img_pth; if (_xml_type == "epub") { _img_pth = "image/"; } else if (_xml_type == "scroll") { _img_pth = "../../image/"; } else if (_xml_type == "seg") { _img_pth = "../../../image/"; } if (_txt.match(rgx.inline_image)) { _txt = replaceAll!(m => m["pre"].to!string ~ _img_element(_img_pth ~ m["img"].to!string, m["width"].to!string, m["height"].to!string, _image_alt(m["post"].to!string)) ~ " " ~ _image_rest(m["post"].to!string) )(_txt, rgx.inline_image); } return _txt; } /+ ↓ the same treatment of an image as the html and epub writers give it, because this is the html the cgi search renders: the markup's description is the alt text, and dimensions that are not there are not written as zeroes. +/ string _image_alt(string _post) { if (auto m = _post.matchFirst(rgx.inline_image_alt)) { return m["alt"].to!string; } return ""; } string _image_rest(string _post) { if (auto m = _post.matchFirst(rgx.inline_image_alt)) { return m.post.to!string; } return _post; } string _img_element(string _src, string _w, string _h, string _alt) { string _dim = (_w == "0" && _h == "0") ? "" : " width=\"" ~ _w ~ "\" height=\"" ~ _h ~ "\""; return "\"""; } #+END_SRC ******* links ******** scroll, seg, epub #+NAME: sanitize_and_munge_inline_html_inline_links #+BEGIN_SRC d string inline_links(M,O)( M doc_matters, const O obj, string _txt, string _xml_type = "seg", ) { if (obj.has.inline_links) { if (obj.metainfo.is_a != "code") { _txt = replaceAll!(m => m["linked_text"] ~ "┤" ~ to!string((obj.stow.link[m["num"].to!ulong])).encode ~ "├" )(_txt, rgx.inline_link_number_only); } if ((_txt.match(rgx.mark_internal_site_lnk)) && (_xml_type == "scroll")) { // conditions reversed to avoid: gdc compiled program run segfault _txt = _txt.replaceAll( rgx.inline_seg_link, "$1"); } auto pth_html = spinePathsHTML!()(doc_matters.output_path, doc_matters.src.language); if (_xml_type == "seg") { foreach (m; _txt.matchAll(rgx.inline_link_seg_and_hash)) { if (m.captures["hash"] in doc_matters.has.tag_associations) { if (m.captures["hash"] == doc_matters.has.tag_associations[(m.captures["hash"])]["seg_lv4"]) { _txt = _txt.replaceFirst( rgx.inline_link_seg_and_hash, "┥$1┝┤" ~ doc_matters.conf_make_meta.conf.w_srv_data_root_url_html ~ "/" ~ pth_html.tail_fn_seg(doc_matters.src.filename, "$2.html") ~ "├" ); } else { _txt = _txt.replaceFirst( rgx.inline_link_seg_and_hash, "┥$1┝┤" ~ doc_matters.conf_make_meta.conf.w_srv_data_root_url_html ~ "/" ~ doc_matters.src.filename_base ~ "/" ~ doc_matters.has.tag_associations[(m.captures["hash"])]["seg_lv4"] ~ ".html" ~ "#" ~ m.captures["hash"] ~ "├" ); } } else { if (doc_matters.opt.action.vox_gt_1) { writeln( "WARNING on internal document links, anchor to link <<" ~ m.captures["hash"] ~ ">> not found in document, " ~ "anchor: " ~ m.captures["hash"] ~ " document: " ~ doc_matters.src.filename ); } } } } else { if (auto m = _txt.match(rgx.inline_link_seg_and_hash)) { _txt = _txt.replaceFirst( rgx.inline_link_seg_and_hash, "┥$1┝┤" ~ doc_matters.conf_make_meta.conf.w_srv_data_root_url_html ~ "/" ~ pth_html.tail_fn_scroll(doc_matters.src.filename) ~ "#" ~ m.captures["hash"] ~ "├" ); } } _txt = _txt .replaceAll( rgx.inline_link_fn_suffix, ("$1.html")) .replaceAll( rgx.inline_link, ("$1")) .replaceAll( rgx.mark_internal_site_lnk, ""); } debug(markup_links) { if (_txt.match(rgx.inline_link)) { writeln(__LINE__, " (missed) markup link identified (", obj.has.inline_links, "): ", obj.metainfo.is_a, ": ", obj.text ); } // if (obj.metainfo.is_a == "bookindex") { // DEBUG LINE // if (_txt.match(regex(r"" ~ "" ~ m.captures["num"] ~ "." ~ m.captures["note"] ~ "

"; } _txt = replaceAll!(m => (" " ~ "" ~ m["num"] ~ "")) (_txt, rgx.inline_notes_al_regular_number_note) ~ _endnotes.join("\n"); } debug(markup_endnotes) { if (_txt.match(rgx.inline_notes_al_regular_number_note)) { writeln(__LINE__, " (missed) markup endnote: ", obj.metainfo.is_a, ": ", obj.text); } } debug(markup) { if (_txt.match(rgx.inline_notes_al_regular_number_note)) { writeln(__LINE__, " (missed) markup endnote: ", obj.metainfo.is_a, ": ", obj.text); } } return _txt; } #+END_SRC ******* inline markup (formatting) #+NAME: sanitize_and_munge_inline_html_inline_markup #+BEGIN_SRC d string xml_type="seg"; /+ set html document type to be linked to here (seg|scroll) +/ string inline_markup(M,O)( M doc_matters, const O obj, string _txt, ) { if (obj.metainfo.is_a == "group") { _txt = inline_grouped_text_bullets_indents(doc_matters, obj, _txt, xml_type); } _txt = inline_images(doc_matters, obj, _txt, xml_type); _txt = inline_links(doc_matters, obj, _txt, xml_type); _txt = inline_notes_scroll(doc_matters, obj, _txt); return _txt; } #+END_SRC ***** objects (formatting) ****** heading #+NAME: html_objects_heading #+BEGIN_SRC d string html_heading(M,O)( M doc_matters, const O obj, ) { assert(obj.metainfo.is_of_part == "body" || "frontmatter" || "backmatter"); assert(obj.metainfo.is_of_section == "body" || "toc" || "endnotes" || "glossary" || "bibliography" || "bookindex" || "blurb"); assert(obj.metainfo.is_of_type == "para"); assert(obj.metainfo.is_a == "heading"); string _txt = munge_html(doc_matters, obj); _txt = inline_markup(doc_matters, obj, _txt); string o = format(q"┃

%s

┃", obj.metainfo.is_a, _txt, ); return o; } #+END_SRC ******* +fancy+ ##+NAME: prepare_objects_html #+BEGIN_SRC d string html_heading(M,O)( M doc_matters, const O obj, ) { string o; string _txt = munge_html(doc_matters, obj); o = format(q"┃ %s ┃", obj.metainfo.heading_lev_markup, obj.metainfo.is_a, _txt, obj.metainfo.heading_lev_markup, ); return o; } #+END_SRC ****** para #+NAME: html_objects_para #+BEGIN_SRC d string html_para(M,O)( M doc_matters, const O obj, ) { assert(obj.metainfo.is_of_part == "body" || "frontmatter" || "backmatter"); assert(obj.metainfo.is_of_section == "body" || "toc" || "endnotes" || "glossary" || "bibliography" || "bookindex" || "blurb"); assert(obj.metainfo.is_of_type == "para"); assert(obj.metainfo.is_a == "para" || "toc" || "endnote" || "glossary" || "bibliography" || "bookindex" || "blurb"); string _txt = munge_html(doc_matters, obj); _txt = (obj.attrib.bullet) ? ("●  " ~ _txt) : _txt; _txt = inline_markup(doc_matters, obj, _txt); string o = format(q"┃

%s

┃", obj.metainfo.is_a, obj.attrib.indent_hang, obj.attrib.indent_base, _txt ); return o; } #+END_SRC ****** quote #+NAME: html_objects_quote #+BEGIN_SRC d string html_quote(M,O)( M doc_matters, const O obj, ) { assert(obj.metainfo.is_of_part == "body"); assert(obj.metainfo.is_of_section == "body" || "glossary" || "bibliography" || "bookindex" || "blurb"); assert(obj.metainfo.is_of_type == "block"); assert(obj.metainfo.is_a == "quote"); string _txt = munge_html(doc_matters, obj); string o = format(q"┃

%s

┃", obj.metainfo.is_a, _txt ); return o; } #+END_SRC ****** group #+NAME: html_objects_group #+BEGIN_SRC d string html_group(M,O)( M doc_matters, const O obj, ) { assert(obj.metainfo.is_of_part == "body"); assert(obj.metainfo.is_of_section == "body" || "glossary" || "bibliography" || "bookindex" || "blurb"); assert(obj.metainfo.is_of_type == "block"); assert(obj.metainfo.is_a == "group"); string _txt = munge_html(doc_matters, obj); _txt = inline_markup(doc_matters, obj, _txt); string o = format(q"┃

%s

┃", obj.metainfo.is_a, _txt ); return o; } #+END_SRC ****** block #+NAME: html_objects_block #+BEGIN_SRC d string html_block(M,O)( M doc_matters, const O obj, ) { assert(obj.metainfo.is_of_part == "body"); assert(obj.metainfo.is_of_section == "body" || "glossary" || "bibliography" || "bookindex" || "blurb"); assert(obj.metainfo.is_of_type == "block"); assert(obj.metainfo.is_a == "block"); string _txt = munge_html(doc_matters, obj); _txt = inline_markup(doc_matters, obj, _txt); string o = format(q"┃

%s

┃", obj.metainfo.is_a, _txt.stripRight ); return o; } #+END_SRC ****** verse #+NAME: html_objects_verse #+BEGIN_SRC d string html_verse(M,O)( M doc_matters, const O obj, ) { assert(obj.metainfo.is_of_part == "body"); assert(obj.metainfo.is_of_section == "body" || "glossary" || "bibliography" || "bookindex" || "blurb"); assert(obj.metainfo.is_of_type == "block"); assert(obj.metainfo.is_a == "verse"); string _txt = munge_html(doc_matters, obj); string o = format(q"┃

%s

┃", obj.metainfo.is_a, _txt ); return o; } #+END_SRC ****** code #+NAME: html_objects_code #+BEGIN_SRC d string html_code(O)( const O obj, ) { assert(obj.metainfo.is_of_part == "body"); assert(obj.metainfo.is_of_section == "body"); assert(obj.metainfo.is_of_type == "block"); assert(obj.metainfo.is_a == "code"); string _txt = html_special_characters_code(obj.text); string o = format(q"┃

%s

┃", obj.metainfo.is_a, _txt ); return o; } #+END_SRC ****** table #+NAME: html_objects_table #+BEGIN_SRC d string html_table(M,O)( M doc_matters, const O obj, ) { assert(obj.metainfo.is_of_part == "body"); assert(obj.metainfo.is_of_section == "body"); assert(obj.metainfo.is_of_type == "block"); assert(obj.metainfo.is_a == "table"); Tuple!(string, string) _tablarize(O)( const O obj, string _txt, ) { string[] _table_rows = _txt.split(rgx.table_delimiter_row); string[] _table_cols; string _table; string _tablenote; bool _in_head = false; bool _in_body = false; foreach(row_idx, row; _table_rows) { _table_cols = row.split(rgx.table_delimiter_col); if ((_table_cols.length == 1) && (_table_rows.length <= row_idx+2)) { // check row_idx+2 (rather than == ++row_idx) _tablenote ~= _table_cols[0]; continue; } /+ ↓ a heading row goes in , the rest in +/ if (row_idx == 0 && obj.table.heading) { _table ~= ""; _in_head = true; } else if (!_in_body) { if (_in_head) { _table ~= ""; _in_head = false; } _table ~= ""; _in_body = true; } _table ~= ""; foreach(col_idx, cell; _table_cols) { bool _is_th = (row_idx == 0 && obj.table.heading); string _col_is = _is_th ? "th" : "td"; string _class = (col_idx < obj.table.column_aligns.length && obj.table.column_aligns[col_idx] == "l") ? "left" : "right"; _table ~= "<" ~ _col_is ~ (_is_th ? " scope=\"col\"" : "") ~ " class=\"" ~ _class ~ "\">"; _table ~= cell; _table ~= ""; } _table ~= ""; } if (_in_head) { _table ~= ""; } else if (_in_body) { _table ~= ""; } /+ ↓ column widths as a , not a width attribute on every cell +/ string _colgroup; if (obj.table.column_widths.length > 0) { string _cols; foreach(w; obj.table.column_widths) { _cols ~= (w > 0) ? "" : ""; } _colgroup = "" ~ _cols ~ ""; } Tuple!(string, string) t = tuple( (_colgroup ~ _table), _tablenote, ); return t; } string _txt = munge_html(doc_matters, obj); Tuple!(string, string) t = _tablarize(obj, _txt); _txt = t[0]; string _note = t[1]; string o = format(q"┃
%s
%s
┃", obj.metainfo.is_a, _txt, (_note.strip.empty) ? "" : ("

" ~ _note.strip ~ "

") ); return o; } #+END_SRC *** 2. hub (sqlite_format_and_load_objects) **** sql related #+NAME: sqlite_load_object_string #+BEGIN_SRC d string sqlite_load_string(M,O)( M doc_matters, const O obj, ) { string o; return o; } #+END_SRC #+NAME: sqlite_load_object_statement #+BEGIN_SRC d string sqlite_statement(O)( const O obj, string _txt, string _html, ) { void _sql_exe(O)( string _sql, ) { writeln(_html); writeln(_sql); } string _sql; return _sql; } #+END_SRC **** heading #+NAME: hub_format_and_sqlite_load_objects_heading #+BEGIN_SRC d string[string] heading(M,O)( M doc_matters, const O obj, ) { string[string] obj_txt = [ "text": generic_munge_sanitize_text_for_search(obj.text), "html": html_heading(doc_matters, obj) ]; { /+ debug +/ if (doc_matters.opt.action.debug_do_sqlite) { debug(sql_txt) { writeln(obj_txt["text"]); } debug(sql_html) { writeln(obj_txt["html"]); } } else { // load sql } } return obj_txt; } #+END_SRC **** para #+NAME: hub_format_and_sqlite_load_objects_para #+BEGIN_SRC d string[string] para(M,O)( M doc_matters, const O obj, ) { string[string] obj_txt = [ "text": generic_munge_sanitize_text_for_search(obj.text), "html": html_para(doc_matters, obj) ]; { /+ debug +/ if (doc_matters.opt.action.debug_do_sqlite) { debug(sql_txt) { writeln(obj_txt["text"]); } debug(sql_html) { writeln(obj_txt["html"]); } } else { // load sql } } return obj_txt; } #+END_SRC **** quote #+NAME: hub_format_and_sqlite_load_objects_quote #+BEGIN_SRC d string[string] quote(M,O)( M doc_matters, const O obj, ) { string[string] obj_txt = [ "text": generic_munge_sanitize_text_for_search(obj.text), "html": html_quote(doc_matters, obj) ]; { /+ debug +/ if (doc_matters.opt.action.debug_do_sqlite) { debug(sql_txt) { writeln(obj_txt["text"]); } debug(sql_html) { writeln(obj_txt["html"]); } } else { // load sql } } return obj_txt; } #+END_SRC **** group #+NAME: hub_format_and_sqlite_load_objects_group #+BEGIN_SRC d string[string] group(M,O)( M doc_matters, const O obj, ) { string[string] obj_txt = [ "text": generic_munge_sanitize_text_for_search(obj.text), "html": html_group(doc_matters, obj) ]; { /+ debug +/ if (doc_matters.opt.action.debug_do_sqlite) { debug(sql_txt) { writeln(obj_txt["text"]); } debug(sql_html) { writeln(obj_txt["html"]); } } else { // load sql } } return obj_txt; } #+END_SRC **** block #+NAME: hub_format_and_sqlite_load_objects_block #+BEGIN_SRC d string[string] block(M,O)( M doc_matters, const O obj, ) { string[string] obj_txt = [ "text": generic_munge_sanitize_text_for_search(obj.text), "html": html_block(doc_matters, obj) ]; { /+ debug +/ if (doc_matters.opt.action.debug_do_sqlite) { debug(sql_txt) { writeln(obj_txt["text"]); } debug(sql_html) { writeln(obj_txt["html"]); } } else { // load sql } } return obj_txt; } #+END_SRC **** verse #+NAME: hub_format_and_sqlite_load_objects_verse #+BEGIN_SRC d string[string] verse(M,O)( M doc_matters, const O obj, ) { string[string] obj_txt = [ "text": generic_munge_sanitize_text_for_search(obj.text), "html": html_verse(doc_matters, obj) ]; { /+ debug +/ if (doc_matters.opt.action.debug_do_sqlite) { debug(sql_txt) { writeln(obj_txt["text"]); } debug(sql_html) { writeln(obj_txt["html"]); } } else { // load sql } } return obj_txt; } #+END_SRC **** code #+NAME: hub_format_and_sqlite_load_objects_code #+BEGIN_SRC d string[string] code(M,O)( M doc_matters, const O obj, ) { string[string] obj_txt = [ "text": generic_munge_sanitize_text_for_search(obj.text), "html": html_code(obj) ]; { /+ debug +/ if (doc_matters.opt.action.debug_do_sqlite) { debug(sql_txt) { writeln(obj_txt["text"]); } debug(sql_html) { writeln(obj_txt["html"]); } } else { // load sql } } return obj_txt; } #+END_SRC **** table #+NAME: hub_format_and_sqlite_load_objects_table #+BEGIN_SRC d string[string] table(M,O)( M doc_matters, const O obj, ) { string[string] obj_txt = [ "text": generic_munge_sanitize_text_for_search(obj.text), "html": html_table(doc_matters, obj) ]; { /+ debug +/ if (doc_matters.opt.action.debug_do_sqlite) { debug(sql_txt) { writeln(obj_txt["text"]); } debug(sql_html) { writeln(obj_txt["html"]); } } else { // load sql } } return obj_txt; } #+END_SRC ** 4. ↻ loop, identify, load - loop template #+NAME: sqlite_objects_loop #+BEGIN_SRC d auto format_and_sqlite_load = SQLiteFormatAndLoadObject!()(doc.matters); string[string] obj_txt; string doc_text; string[] _insert_doc_objects; foreach (part; doc.matters.has.keys_seq.sql) { foreach (obj; doc.abstraction[part]) { switch (obj.metainfo.is_of_part) { case "frontmatter": assert(part == "head", part); switch (obj.metainfo.is_of_type) { case "para": switch (obj.metainfo.is_a) { case "heading": obj_txt = format_and_sqlite_load.heading(doc.matters, obj); break; default: { /+ debug +/ if (doc.matters.opt.action.debug_do_sqlite) { writeln(__FILE__, ":", __LINE__, ": ", obj.metainfo.is_a); } } break; } break; default: { /+ debug +/ if (doc.matters.opt.action.debug_do_sqlite) { writeln(__FILE__, ":", __LINE__, ": ", obj.metainfo.is_of_type); } } break; } break; case "body": // assert(part == "body", part); switch (obj.metainfo.is_of_type) { case "para": switch (obj.metainfo.is_a) { case "heading": debug (asserts) { if (part != "body") { writeln(__LINE__, ": ", obj.text); } } obj_txt = format_and_sqlite_load.heading(doc.matters, obj); break; case "para": obj_txt = format_and_sqlite_load.para(doc.matters, obj); break; default: { /+ debug +/ if (doc.matters.opt.action.debug_do_sqlite) { writeln(__FILE__, ":", __LINE__, ": ", obj.metainfo.is_a); } } break; } break; case "block": switch (obj.metainfo.is_a) { case "quote": obj_txt = format_and_sqlite_load.quote(doc.matters, obj); break; case "group": obj_txt = format_and_sqlite_load.group(doc.matters, obj); break; case "block": obj_txt = format_and_sqlite_load.block(doc.matters, obj); break; case "poem": // double check on keeping both poem & verse break; case "verse": obj_txt = format_and_sqlite_load.verse(doc.matters, obj); break; case "code": obj_txt = format_and_sqlite_load.code(doc.matters, obj); break; case "table": obj_txt = format_and_sqlite_load.table(doc.matters, obj); break; default: { /+ debug +/ if (doc.matters.opt.action.debug_do_sqlite) { writeln(__FILE__, ":", __LINE__, ": ", obj.metainfo.is_a); } } break; } break; default: { /+ debug +/ if (doc.matters.opt.action.debug_do_sqlite) { writeln(__FILE__, ":", __LINE__, ": ", obj.metainfo.is_of_type); } } break; } break; case "backmatter": assert(part == "glossary" || "bibliography" || "bookindex" || "blurb" || "tail", part); switch (obj.metainfo.is_of_type) { case "para": switch (obj.metainfo.is_a) { case "heading": obj_txt = format_and_sqlite_load.heading(doc.matters, obj); break; case "glossary": assert(part == "glossary", part); obj_txt = format_and_sqlite_load.para(doc.matters, obj); break; case "bibliography": assert(part == "bibliography", part); obj_txt = format_and_sqlite_load.para(doc.matters, obj); break; case "bookindex": assert(part == "bookindex", part); obj_txt = format_and_sqlite_load.para(doc.matters, obj); break; case "blurb": assert(part == "blurb", part); obj_txt = format_and_sqlite_load.para(doc.matters, obj); break; default: { /+ debug +/ if (doc.matters.opt.action.debug_do_sqlite) { writeln(__FILE__, ":", __LINE__, ": ", obj.metainfo.is_a); } } break; } break; default: { /+ debug +/ if (doc.matters.opt.action.debug_do_sqlite) { writeln(__FILE__, ":", __LINE__, ": ", obj.metainfo.is_of_type); } } break; } break; case "comment": break; default: { /+ debug +/ if (doc.matters.opt.action.debug_do_sqlite) { writeln(__FILE__, ":", __LINE__, ": ", obj.metainfo.is_of_part); // check where empty value could come from writeln(__FILE__, ":", __LINE__, ": ", obj.metainfo.is_a); writeln(__FILE__, ":", __LINE__, ": ", obj.text); // check where empty value could come from } } break; } if (obj.metainfo.is_a == "heading") { if (doc.matters.opt.action.show_sqlite) { if (obj.metainfo.heading_lev_markup == 0) { writeln(doc.matters.src.filename); } writeln( "markup: ", obj.metainfo.heading_lev_markup, "> ", obj.metainfo.dom_structure_markedup_tags_status, "; collapsed: ", obj.metainfo.heading_lev_collapsed, "> ", obj.metainfo.dom_structure_collapsed_tags_status, "; ocn: ", obj.metainfo.ocn, "; parent: ", obj.metainfo.parent_lev_markup, "; ocn: ", obj.metainfo.parent_ocn, "; ", ); } } if (!(obj.metainfo.is_a == "comment")) { _insert_doc_objects ~= insertDocObjectsRow(obj); } } // loop closes } return _insert_doc_objects.join.to!(char[]).toUTF8; #+END_SRC ** 5. SQL statements :statement: *** drop index and tables **** DROP INDEX IF EXISTS #+NAME: sqlite_statement_drop_existing_index #+BEGIN_SRC sql DROP INDEX IF EXISTS idx_ocn; DROP INDEX IF EXISTS idx_uid; DROP INDEX IF EXISTS idx_digest_clean; DROP INDEX IF EXISTS idx_digest_all; DROP INDEX IF EXISTS idx_clean; DROP INDEX IF EXISTS idx_title; DROP INDEX IF EXISTS idx_author; DROP INDEX IF EXISTS src_filename_base; DROP INDEX IF EXISTS idx_language_document_char; DROP INDEX IF EXISTS idx_classify_topic_register; DROP INDEX IF EXISTS idx_topic_list; #+END_SRC **** DROP TABLE IF EXISTS #+NAME: sqlite_statement_drop_existing_tables #+BEGIN_SRC sql DROP TABLE IF EXISTS metadata_and_text; DROP TABLE IF EXISTS topic_register; DROP TABLE IF EXISTS doc_objects; DROP TABLE IF EXISTS urls; #+END_SRC *** create tables and index **** CREATE TABLE IF NOT EXISTS metadata_and_text #+NAME: sqlite_statement_create_table_metadata_and_src_txt #+BEGIN_SRC sql CREATE TABLE IF NOT EXISTS metadata_and_text ( uid VARCHAR(256) UNIQUE, -- filename, language char, pod/txt (decide on delimiter [,;:/]) src_composite_id_per_txt VARCHAR(256) NOT NULL, -- UNIQUE, z pod name if any + src filename + language code src_composite_id_per_pod VARCHAR(256) NOT NULL, -- z pod name if any + src filename title VARCHAR(800) NOT NULL, title_main VARCHAR(400) NOT NULL, title_sub VARCHAR(400) NULL, title_short VARCHAR(400) NULL, title_edition VARCHAR(10) NULL, title_language VARCHAR(100) NULL, title_language_char VARCHAR(6) NULL, creator_author VARCHAR(600) NOT NULL, creator_author_last_first VARCHAR(600) NOT NULL, creator_author_email VARCHAR(100) NULL, creator_author_hon VARCHAR(100) NULL, creator_author_nationality VARCHAR(100) NULL, creator_editor VARCHAR(600) NULL, creator_contributor VARCHAR(600) NULL, creator_illustrator VARCHAR(600) NULL, creator_photographer VARCHAR(600) NULL, creator_translator VARCHAR(600) NULL, creator_prepared_by VARCHAR(600) NULL, creator_digitized_by VARCHAR(600) NULL, creator_audio VARCHAR(600) NULL, creator_video VARCHAR(600) NULL, language_document VARCHAR(100) NULL, language_document_char VARCHAR(6) NOT NULL, language_original VARCHAR(100) NULL, language_original_char VARCHAR(6) NULL, date_added_to_site VARCHAR(10) NULL, date_available VARCHAR(10) NULL, date_created VARCHAR(10) NULL, date_issued VARCHAR(10) NULL, date_modified VARCHAR(10) NULL, date_published VARCHAR(10) NULL, date_valid VARCHAR(10) NULL, date_translated VARCHAR(10) NULL, date_original_publication VARCHAR(10) NULL, date_generated VARCHAR(10) NULL, original_title VARCHAR(800) NULL, original_publisher VARCHAR(600) NULL, original_language VARCHAR(100) NULL, original_language_char VARCHAR(6) NULL, original_source VARCHAR(600) NULL, original_institution VARCHAR(600) NULL, original_nationality VARCHAR(100) NULL, rights_copyright VARCHAR(2500) NULL, rights_copyright_audio VARCHAR(2500) NULL, rights_copyright_cover VARCHAR(2500) NULL, rights_copyright_illustrations VARCHAR(2500) NULL, rights_copyright_photographs VARCHAR(2500) NULL, rights_copyright_text VARCHAR(2500) NULL, rights_copyright_translation VARCHAR(2500) NULL, rights_copyright_video VARCHAR(2500) NULL, rights_license VARCHAR(2500) NULL, identifier_oclc VARCHAR(30) NULL, identifier_isbn VARCHAR(16) NULL, classify_topic_register VARCHAR(2500) NULL, classify_subject VARCHAR(600) NULL, classify_loc VARCHAR(30) NULL, classify_dewey VARCHAR(30) NULL, classify_keywords VARCHAR(600) NULL, notes_abstract TEXT NULL, notes_description TEXT NULL, notes_comment TEXT NULL, notes_coverage VARCHAR(200) NULL, notes_relation VARCHAR(200) NULL, notes_history VARCHAR(600) NULL, notes_type VARCHAR(600) NULL, notes_format VARCHAR(600) NULL, notes_prefix TEXT NULL, notes_prefix_a TEXT NULL, notes_prefix_b TEXT NULL, notes_suffix TEXT NULL, publisher VARCHAR(600) NULL, src_filename_base VARCHAR(256) NOT NULL, src_filename_suffix VARCHAR(6) NOT NULL, src_fingerprint VARCHAR(256) NULL, src_filesize VARCHAR(10) NULL, src_wordcount VARCHAR(10) NULL, pod_name VARCHAR(256) NULL, -- zipped pod, work to be done here pod_fingerprint VARCHAR(256) NULL, -- zipped pod, work to be done here pod_size VARCHAR(10) NULL, -- zipped pod, work to be done here site_url_doc_root VARCHAR(256) NULL, -- url path to doc root site_url_html_toc VARCHAR(256) NULL, site_url_html_scroll VARCHAR(256) NULL, site_url_epub VARCHAR(256) NULL, links TEXT NULL ); #+END_SRC **** CREATE TABLE IF NOT EXISTS topic_register #+NAME: sqlite_statement_create_table_topic_register #+BEGIN_SRC sql CREATE TABLE IF NOT EXISTS topic_register ( -- tid BIGINT PRIMARY KEY, uid_metadata_and_text VARCHAR(256) REFERENCES metadata_and_text(uid) ON DELETE CASCADE, -- src_composite_id_per_txt VARCHAR(256) NOT NULL, - UNIQUE, - z pod name if any + src filename + language code -- src_composite_id_per_pod VARCHAR(256) NOT NULL, - z pod name if any + src filename topic_register VARCHAR(250) NOT NULL, site_url_doc_root VARCHAR(256) NULL, -- url path to doc root site_url_html_toc VARCHAR(256) NULL, site_url_html_scroll VARCHAR(256) NULL ); #+END_SRC **** CREATE TABLE site_urls ? #+NAME: sqlite_statement_create_table_site_urls #+BEGIN_SRC sql CREATE TABLE IF NOT EXISTS site_urls ( -- tid BIGINT PRIMARY KEY, uid_metadata_and_text VARCHAR(256) REFERENCES metadata_and_text(uid) ON DELETE CASCADE, src_composite_id_per_txt VARCHAR(256) NOT NULL, -- UNIQUE, - z pod name if any + src filename + language code src_composite_id_per_pod VARCHAR(256) NOT NULL, -- z pod name if any + src filename site_url_doc_root VARCHAR(256) NULL, -- url path to doc root site_url_html_toc VARCHAR(256) NULL, site_url_html_scroll VARCHAR(256) NULL ); #+END_SRC **** CREATE TABLE doc_objects #+NAME: sqlite_statement_create_table_objects #+BEGIN_SRC sql CREATE TABLE IF NOT EXISTS doc_objects ( lid BIGINT PRIMARY KEY, uid_metadata_and_text VARCHAR(256) REFERENCES metadata_and_text(uid) ON DELETE CASCADE, ocn SMALLINT, obj_id VARCHAR(6) NULL, clean TEXT NULL, body TEXT NULL, seg VARCHAR(256) NULL, lev_an VARCHAR(1), is_of_type VARCHAR(16), is_a VARCHAR(16), lev SMALLINT NULL, node VARCHAR(16) NULL, -- not written to parent VARCHAR(16) NULL, last_descendant VARCHAR(16) NULL, -- headings only digest_clean CHAR(256), digest_all CHAR(256), seg_name CHAR(256), types CHAR(1) NULL ); #+END_SRC **** CREATE INDEX #+NAME: sqlite_statement_create_index #+BEGIN_SRC sql CREATE INDEX IF NOT EXISTS idx_ocn ON doc_objects(ocn); CREATE INDEX IF NOT EXISTS idx_digest_clean ON doc_objects(digest_clean); CREATE INDEX IF NOT EXISTS idx_digest_all ON doc_objects(digest_all); CREATE INDEX IF NOT EXISTS idx_clean ON doc_objects(clean); CREATE INDEX IF NOT EXISTS idx_title ON metadata_and_text(title); CREATE INDEX IF NOT EXISTS idx_author ON metadata_and_text(creator_author_last_first); CREATE INDEX IF NOT EXISTS idx_uid ON metadata_and_text(uid); CREATE INDEX IF NOT EXISTS idx_filename ON metadata_and_text(src_filename_base); CREATE INDEX IF NOT EXISTS idx_language ON metadata_and_text(language_document_char); CREATE INDEX IF NOT EXISTS idx_topics ON metadata_and_text(classify_topic_register); CREATE INDEX IF NOT EXISTS idx_topic_list ON topic_register(topic_register); #+END_SRC *** TODO local site link & info *** delete rows (delete document) **** DELETE uid rows doc matters & metadata ***** sql statement: dlang format #+NAME: sqlite_formatted_delete_format #+BEGIN_SRC d string _uid = doc_matters.src.doc_uid; string _delete_uid = format(q"┃ #+END_SRC ***** DELETE FROM ... WHERE #+NAME: sqlite_formatted_delete_sql #+BEGIN_SRC sql DELETE FROM metadata_and_text WHERE uid = '%s'; DELETE FROM doc_objects WHERE uid_metadata_and_text = '%s'; #+END_SRC ***** VALUES #+NAME: sqlite_formatted_delete_values #+BEGIN_SRC d ┃", _uid, _uid, ); #+END_SRC *** inserts **** INSERT doc matters & metadata ***** sql statement: dlang format #+NAME: sqlite_formatted_insertions_doc_matters_metadata_format #+BEGIN_SRC d string _uid = SQLinsertDelimiter!()(doc_matters.src.doc_uid); string _insert_metadata = format(q"┃ #+END_SRC ***** INSERT INTO #+NAME: sqlite_formatted_insertions_doc_matters_metadata_sql #+BEGIN_SRC sql INSERT INTO metadata_and_text ( uid, src_filename_base, src_filename_suffix, src_composite_id_per_txt, src_composite_id_per_pod, title, title_main, title_sub, title_short, title_edition, title_language, creator_author, creator_author_last_first, creator_author_email, creator_illustrator, creator_translator, language_document, language_document_char, date_added_to_site, date_available, date_created, date_issued, date_modified, date_published, date_valid, rights_copyright, rights_copyright_audio, rights_copyright_cover, rights_copyright_illustrations, rights_copyright_photographs, rights_copyright_text, rights_copyright_translation, rights_copyright_video, rights_license, identifier_oclc, identifier_isbn, classify_dewey, classify_keywords, classify_loc, classify_subject, classify_topic_register, original_title, original_publisher, original_language, original_language_char, original_source, notes_abstract, notes_description, publisher, site_url_doc_root ) #+END_SRC ***** VALUES #+NAME: sqlite_formatted_insertions_doc_matters_metadata_sql_values #+BEGIN_SRC sql VALUES ( '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s', '%s' ); #+END_SRC ***** dlang values for formatting #+NAME: sqlite_formatted_insertions_doc_matters_metadata_values #+BEGIN_SRC d ┃", _uid, SQLinsertDelimiter!()(doc_matters.src.filename_base), SQLinsertDelimiter!()(doc_matters.src.filename_extension), SQLinsertDelimiter!()(doc_matters.src.docname_composite_unique_per_src_doc), SQLinsertDelimiter!()(doc_matters.src.docname_composite_unique_per_src_pod), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.title_full), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.title_main), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.title_subtitle), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.title_short), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.title_edition), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.title_language), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.creator_author), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.creator_author_surname_fn), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.creator_author_email), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.creator_illustrator), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.creator_translator), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.language_document), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.language_document_char), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.date_added_to_site), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.date_available), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.date_created), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.date_issued), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.date_modified), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.date_published), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.date_valid), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.rights_copyright), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.rights_copyright_audio), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.rights_copyright_cover), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.rights_copyright_illustrations), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.rights_copyright_photographs), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.rights_copyright_text), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.rights_copyright_translation), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.rights_copyright_video), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.rights_license), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.identifier_oclc), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.identifier_isbn), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.classify_dewey), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.classify_keywords), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.classify_loc), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.classify_subject), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.classify_topic_register), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.notes_abstract), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.notes_description), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.original_title), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.original_publisher), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.original_language), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.original_language_char), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.original_source), SQLinsertDelimiter!()(doc_matters.conf_make_meta.meta.publisher), SQLinsertDelimiter!()(doc_matters.conf_make_meta.conf.w_srv_data_root_url_html) ); #+END_SRC **** INSERT topic register writeln(doc_matters.conf_make_meta.meta.classify_topic_register_arr); ***** loop sql statement: dlang format #+NAME: sqlite_formatted_insertions_topic_register_loop #+BEGIN_SRC d string _uid = SQLinsertDelimiter!()(doc_matters.src.doc_uid); string[] _insert_topics; foreach (topic_line; doc_matters.conf_make_meta.meta.classify_topic_register_expanded_arr) { // writeln(topic_line); #+END_SRC ***** sql statement: dlang format #+NAME: sqlite_formatted_insertions_topic_register_format #+BEGIN_SRC d _insert_topics ~= format(q"┃ #+END_SRC ***** INSERT INTO #+NAME: sqlite_formatted_insertions_topic_register_sql #+BEGIN_SRC sql INSERT INTO topic_register ( uid_metadata_and_text, topic_register ) #+END_SRC ***** VALUES #+NAME: sqlite_formatted_insertions_topic_register_sql_values #+BEGIN_SRC sql VALUES ( '%s', '%s' ); #+END_SRC ***** dlang values for formatting #+NAME: sqlite_formatted_insertions_topic_register_values #+BEGIN_SRC d ┃", _uid, SQLinsertDelimiter!()(topic_line) ); #+END_SRC **** INSERT doc objects lid unique, increment by 1 per object, not ocn metadata tid document number unique either: - increment by adding 1 for each document, - make hash of document filename or url and use? ***** sql statement: dlang format #+NAME: sqlite_formatted_insertions_doc_objects_format #+BEGIN_SRC d string _insert_doc_objects_row = format(q"┃ #+END_SRC ***** INSERT INTO #+NAME: sqlite_formatted_insertions_doc_objects_sql #+BEGIN_SRC sql INSERT INTO doc_objects ( uid_metadata_and_text, ocn, obj_id, clean, body, lev, is_of_type, is_a, seg_name ) #+END_SRC ***** VALUES #+NAME: sqlite_formatted_insertions_doc_objects_sql_values #+BEGIN_SRC sql VALUES ( '%s', %s, '%s', '%s', '%s', %s, '%s', '%s', '%s' ); #+END_SRC ***** dlang values for formatting #+NAME: sqlite_formatted_insertions_doc_objects_values #+BEGIN_SRC d ┃", _uid, obj.metainfo.ocn, obj.metainfo.identifier, SQLinsertDelimiter!()(obj_txt["text"]), SQLinsertDelimiter!()(obj_txt["html"]), obj.metainfo.heading_lev_markup, obj.metainfo.is_of_type, obj.metainfo.is_a, obj.tags.html_segment_anchor_tag_is ); #+END_SRC * org includes * sqlite collection search db schema version #+NAME: sqlite_collection_search_db_schema_version #+BEGIN_SRC org <<./sisudoc_spine_version_info_and_doc_header_including_copyright_and_license.org:sqlite_collection_search_db_schema_version()>> #+END_SRC ** project version #+NAME: spine_version #+HEADER: :noweb yes #+BEGIN_SRC emacs-lisp <<./sisudoc_spine_version_info_and_doc_header_including_copyright_and_license.org:spine_project_version()>> #+END_SRC ** year #+NAME: year #+HEADER: :noweb yes #+BEGIN_SRC emacs-lisp <<./sisudoc_spine_version_info_and_doc_header_including_copyright_and_license.org:year()>> #+END_SRC ** document header including copyright & license #+NAME: doc_header_including_copyright_and_license #+HEADER: :noweb yes #+BEGIN_SRC emacs-lisp <<./sisudoc_spine_version_info_and_doc_header_including_copyright_and_license.org:spine_doc_header_including_copyright_and_license()>> #+END_SRC * __END__