From 84428c144e54170ec629b379e1daf86857da2be8 Mon Sep 17 00:00:00 2001 From: Maël Gassmann Date: Thu, 17 Mar 2022 16:14:17 +0100 Subject: [~] Refactoring of the FileAssembler, Extraction of methods in PParser --- src/CMakeLists.txt | 1 + src/FileAssembler.cpp | 165 ++++------------------------------- src/PParser.cpp | 236 ++++++++++++++++++++++++++++++++++++++++++++++++++ src/main.cpp | 53 +++++------- 4 files changed, 274 insertions(+), 181 deletions(-) create mode 100644 src/PParser.cpp (limited to 'src') diff --git a/src/CMakeLists.txt b/src/CMakeLists.txt index 81f1e7d..eac87bf 100644 --- a/src/CMakeLists.txt +++ b/src/CMakeLists.txt @@ -4,6 +4,7 @@ add_executable( ${EXE} main.cpp FileAssembler.cpp + PParser.cpp ) add_custom_target(run COMMAND ${EXE} diff --git a/src/FileAssembler.cpp b/src/FileAssembler.cpp index 7a64846..3949ca7 100644 --- a/src/FileAssembler.cpp +++ b/src/FileAssembler.cpp @@ -1,12 +1,9 @@ #include "../inc/FileAssembler.h" +#include "../inc/PParser.h" #include "../inc/maddy/parser.h" #include -#include -#include #include #include -#include -#include using namespace std; namespace fs = filesystem; @@ -40,7 +37,6 @@ FileAssembler::FileAssembler(string path): path(path){ } } } - ordered_posts_indexes = get_ordered_posts_indexes(); // Only used when a listing of posts in a page is necessary } void FileAssembler::parse_variables(){ @@ -77,154 +73,25 @@ string FileAssembler::get_file_content(string path){ return content; } -map FileAssembler::get_pages(){ - return assemble_from_iterator(pages.begin(), pages.end(), false); +map* FileAssembler::get_website(){ + PParser p = PParser(&variables, &templates, &pages, &posts); + return p.parse(); } -map FileAssembler::get_posts(){ - return assemble_from_iterator(posts.begin(), posts.end(), true); -} -map FileAssembler::assemble_from_iterator(map::iterator it, map::iterator end, bool is_post){ - if(templates.find("header.html") == templates.end()){ - cerr << "Error: swg: header.html is not present in the sourced folder." << endl; - exit(2); - } else if(templates.find("footer.html") == templates.end()){ - cerr << "Error: swg: footer.html is not present in the sourced folder." << endl; - exit(2); - } - map p_it; - while (it != end){ - if(it->first.substr(0, 5) != "link_"){ // Ignoring link pages - p_it[it->first] = parse(it->first, templates["header.html"] + it->second + templates["footer.html"], is_post); - } - it ++; - } - return p_it; +string FileAssembler::get_target(){ + string w = variables["website"]; + static std::regex rgx("\\w+\\.\\w+"); + std::smatch match; + if (std::regex_search(w, match, rgx)) + return match[0]; + cerr << "Error: swg: website attribute is badly configured!" << endl; + exit(7); } -string FileAssembler::parse(string title, string to_parse, bool is_post){ - string parsed = to_parse; - string url = variables["website"]; - - if(is_post) - variables["link"] = url + "posts/" + lowercase(title) + ".html"; - else{ - if(title == variables["index"]){ - variables["link"] = url + "index.html"; - }else if(title.substr(0, 5) == "link_"){ // Link - variables["link"] = pages[title]; - }else{ - variables["link"] = url + lowercase(title) + ".html"; - } - variables.erase("date"); // we are not doing it if it's a post, because it could be a listing - } - // Parsing variables and functions - size_t pos_first = 0; - size_t pos_second = 0; - static std::regex r_expression("\\$([^\\$]*)\\$"); - std::smatch m_expression; - while (regex_search(parsed, m_expression, r_expression)){ - string input = m_expression[1]; //Group 1 ie. without the dollars - string output = ""; - - if(input == "title") //TITLE - if(title.substr(0, 5) == "link_") // Link - output = title.substr(5,title.length()-5); - else - output = title; - - else if(input.substr(0, 5) == "date("){ - variables["date"] = parse_arg(input); - output = variables["date"]; - }else if(variables.find(input) != variables.end()) //VARIABLES - output = variables[input]; - - else if(input.substr(0,4) == "res("){ //RESOURCES - string full_path = input.substr(4,input.length()-1-4); - output = url + full_path; - - if(full_path.find("/") != string::npos){ - size_t s; - - string tok; - string last_tok; - - while ((s = full_path.find("/")) != string::npos){ // Adding the path progressively - tok = full_path.substr(0,s-1); - if(find(cached_res.begin(), cached_res.end(), last_tok+tok) == cached_res.end()) - cached_res.push_back(last_tok+tok); // Not found, adding it - last_tok += tok+"/"; - full_path.erase(0,s); - } - } - if(find(cached_res.begin(), cached_res.end(), input.substr(4,input.length()-1-4)) == cached_res.end()) - cached_res.push_back(input.substr(4,input.length()-1-4)); // Not found, adding it - - - } else if(input.substr(0,5) == "list_"){ //LISTINGS - string name = ""; - int arg = -1; - static std::regex r_listing("\\_([a-z]*)(\\((\\d*)\\))?"); - std::smatch m_listing; - if (std::regex_search(input, m_listing, r_listing)){ - name = m_listing[1]; - if(m_listing[3] != "") - arg = stoi(m_listing[3]); - } - string list_template = name + "_listing.html"; - - if(templates.find(list_template) == templates.end()){ - cerr << "Error: swg: Listing template '" << list_template << "' does not exist." << endl; - exit(2); - } - - if(name == "menu"){ - string current_link = variables["link"]; - map::iterator it = pages.begin(); - while (it != pages.end()){ - output += parse(it->first,templates[list_template], false); - it ++; - } - variables["link"] = current_link; - }else if(name == "post"){ - string current_date = ""; - string current_link = variables["link"]; - if(variables.find("date") != templates.end()) - current_date = variables["date"]; - - int i = 0; - vector::iterator it = ordered_posts_indexes.begin(); - while (it != ordered_posts_indexes.end() && - (arg == -1 || i < arg) // Limiting the number of posts shown if arg is set - ){ - string date = parse_arg(posts[*it]); - if(date != "") - variables["date"] = date; - else{ - cerr << "Error: swg: Variable 'date' of post '" << *it << "' is not defined." << endl; - exit(5); - } - output += parse(*it,templates[list_template], true); - it ++; - i ++; - } - if(current_date != "") - variables["date"] = current_date; - else - variables.erase("date"); - variables["link"] = current_link; - } - } - if(output.length() == 0){ - cerr << "Error: swg: Invalid swg text section: \"" << input << "\"." << endl; - exit(4); - } - //cout << "BEFORE:" << endl << parsed << endl << endl; - parsed = regex_replace(parsed, r_expression, output, regex_constants::format_first_only); - //cout << "AFTER:" << endl << parsed << endl ; - //cout << endl << "----------------------------------------------------" << endl << endl; +string FileAssembler::get_index(){ + if(variables.find("index") == variables.end()){ + return ""; } - return parsed; + return variables["index"]; } - diff --git a/src/PParser.cpp b/src/PParser.cpp new file mode 100644 index 0000000..b4f478c --- /dev/null +++ b/src/PParser.cpp @@ -0,0 +1,236 @@ +#include "../inc/PParser.h" + +PParser::PParser( + map *variables, + map *templates, + map *pages, + map *posts + ): variables(variables), + templates(templates), + pages(pages), + posts(posts) +{ + ordered_posts_indexes = get_ordered_posts_indexes(); // Only used when a listing of posts in a page is necessary +} + +map* PParser::parse(){ + map* website = new map[2]; + + map::iterator it = pages->begin(); + while (it != pages->end()){ + if(it->first.substr(0, 5) != "link_"){ // Ignoring link pages + website[0][it->first] = parse_text(it->first, (*templates)["header.html"] + it->second + (*templates)["footer.html"], false); + } + it ++; + } + + it = posts->begin(); + while (it != posts->end()){ + website[1][it->first] = parse_text(it->first, (*templates)["header.html"] + it->second + (*templates)["footer.html"], true); + it ++; + } + return website; +} + +string* PParser::parse_function(string to_parse){ + static std::regex rgx("(\\w+)\\(([^\\)\\$]*)\\)"); + std::smatch match; + if (std::regex_search(to_parse, match, rgx)){ + string* r = new string[2]; + r[0] = match[1].str(); + r[1] = match[2].str(); + return r; + } + return NULL; +} + +string* PParser::parse_separator(string to_parse, string separator){ + static std::regex rgx("(\\w+)" + separator + "(\\w+)"); + std::smatch match; + if (std::regex_search(to_parse, match, rgx)){ + string* r = new string[2]; + r[0] = match[1].str(); + r[1] = match[2].str(); + return r; + } + return NULL; +} + +bool PParser::cmp_posts(pair& a, pair& b){ + string a_date = parse_function(a.second)[1]; + string b_date = parse_function(b.second)[1]; + if(a_date == ""){ + cerr << "Error: swg: Variable 'date' of post '" << a.first << "' is not defined." << endl; + exit(5); + }else if (b_date == ""){ + cerr << "Error: swg: Variable 'date' of post '" << b.first << "' is not defined." << endl; + exit(5); + } + static std::regex rgx("(\\d{1,2})\\.(\\d{1,2})\\.(\\d{4})"); + std::smatch a_match; + std::smatch b_match; + if (std::regex_search(a_date, a_match, rgx)){ // For now only european swiss format handled + if (std::regex_search(b_date, b_match, rgx)){ // For now only european swiss format handled + if(a_match[3] != b_match[3]) // Trying to differentiate by year + return stoi(a_match[3]) > stoi(b_match[3]); + if(a_match[2] != b_match[2]) // Trying to differentiate by month + return stoi(a_match[2]) > stoi(b_match[2]); + if(a_match[1] != b_match[1]) // Trying to differentiate by day + return stoi(a_match[1]) > stoi(b_match[1]); + return false; // or if equal return false + }else{ + cerr << "Error: swg: Variable 'date' of post '" << b.first << "' is not following the format dd.mm.yyyy." << endl; + exit(5); + } + }else{ + cerr << "Error: swg: Variable 'date' of post '" << a.first << "' is not following the format dd.mm.yyyy." << endl; + exit(5); + } +} + +vector PParser::get_ordered_posts_indexes(){ + vector> ordered_posts; + for (auto& it : *posts) { + ordered_posts.push_back(it); + } + sort(ordered_posts.begin(), ordered_posts.end(), cmp_posts); + + vector titles; + transform(ordered_posts.begin(), ordered_posts.end(), std::back_inserter(titles), + [](const std::pair& p) { return p.first; }); + return titles; +} + + + +string PParser::parse_text(string title, string to_parse, bool is_post){ + string parsed = to_parse; + string url = (*variables)["website"]; + + if(is_post) + (*variables)["link"] = url + "posts/" + lowercase(title) + ".html"; + else{ + if(title == (*variables)["index"]){ + (*variables)["link"] = url + "index.html"; + }else if(title.substr(0, 5) == "link_"){ // Link + (*variables)["link"] = (*pages)[title]; + }else{ + (*variables)["link"] = url + lowercase(title) + ".html"; + } + (*variables).erase("date"); // we are not doing it if it's a post, because it could be a listing + } + // Parsing variables and functions + size_t pos_first = 0; + size_t pos_second = 0; + static std::regex swg_expression("\\$([^\\$]*)\\$"); + std::smatch m_swg_expression; + while (regex_search(parsed, m_swg_expression, swg_expression)){ + string input = m_swg_expression[1]; //Group 1 ie. without the dollars + string output = ""; + + if(string *separator = parse_separator(input, "_")) // LISTINGS + { // Can have 0 or 1 argument + if(separator[0] == "list") + { + cout << "Type: " << separator[0] << ", Variation: " << separator[1] << endl; + + int arg = -1; + string *function = parse_function(input); + if (function != NULL) + arg = stoi(function[1]); + + string list_template = separator[1]+ "_listing.html"; + + if((*templates).find(list_template) == (*templates).end()) + { + cerr << "Error: swg: Listing template '" << list_template << "' does not exist." << endl; + exit(2); + } + + if(separator[1] == "menu") + { + string current_link = (*variables)["link"]; + map::iterator it = (*pages).begin(); + while (it != (*pages).end()) + { + output += parse_text(it->first,(*templates)[list_template], false); + it ++; + } + (*variables)["link"] = current_link; + } + else if(separator[1] == "post") + { + string current_date = ""; + string current_link = (*variables)["link"]; + if((*variables).find("date") != (*templates).end()) + current_date = (*variables)["date"]; + + int i = 0; + vector::iterator it = ordered_posts_indexes.begin(); + while (it != ordered_posts_indexes.end() && + (arg == -1 || i < arg)) // Limiting the number of posts shown if arg is set + { + string date = parse_function((*posts)[*it])[1]; + if(date != "") + (*variables)["date"] = date; + else + { + cerr << "Error: swg: Variable 'date' of post '" << *it << "' is not defined." << endl; + exit(5); + } + output += parse_text(*it,(*templates)[list_template], true); + it ++; + i ++; + } + if(current_date != "") + (*variables)["date"] = current_date; + else + (*variables).erase("date"); + (*variables)["link"] = current_link; + } + } + } + else if(string *function = parse_function(input)) // FUNCTIONS + { // *always* have *ONE* argument + cout << "Function: " << function[0] << ", Arg: " << function[1] << endl; + + if(function[0] == "date") // date + { + (*variables)["date"] = function[1]; + output = function[1]; + } + else if(function[0] == "res") // resources + { + string full_path = function[1]; + output = url + full_path; + } + } + else // KEYWORDS + { + cout << "Input: " << input << endl; + + if(input == "title") // Title + { + string* s = parse_separator(title, "_"); + if(s != NULL && s[0] == "link") // Link + output = s[1]; + else + output = title; + } + else if((*variables).find(input) != (*variables).end()) //Variables + output = (*variables)[input]; + + } + + if(output.length() == 0){ + cerr << "Error: swg: Invalid swg text section: \"" << input << "\"." << endl; + exit(4); + } + //cout << "BEFORE:" << endl << parsed << endl << endl; + parsed = regex_replace(parsed, swg_expression, output, regex_constants::format_first_only); + //cout << "AFTER:" << endl << parsed << endl ; + //cout << endl << "----------------------------------------------------" << endl << endl; + } + return parsed; +} + diff --git a/src/main.cpp b/src/main.cpp index 634d6fc..cb5ad2f 100644 --- a/src/main.cpp +++ b/src/main.cpp @@ -1,9 +1,7 @@ #include "../inc/FileAssembler.h" #include -#include #include -#include -#include +#include using namespace std; @@ -21,6 +19,14 @@ void write_file(string path, string content){ } } +static string lowercase(string s){ + string low = ""; + for (int i = 0; i < s.length(); i++) { + low += tolower(s[i]); + } + return low; +} + int generateWebsite(string arg, string config){ // cout << arg << " " << config << endl; if(config[config.length()-1] != '/') @@ -28,8 +34,7 @@ int generateWebsite(string arg, string config){ FileAssembler *fa = new FileAssembler(config); /* TODO * clean or don't clean*/ - map pages = fa->get_pages(); - map posts = fa->get_posts(); + map* website = fa->get_website(); string target = fa->get_target(); string index = fa->get_index(); @@ -41,43 +46,27 @@ int generateWebsite(string arg, string config){ cerr << "Error: swg: Directory '" << target << "/posts' could not be created!" << endl; exit(6); } - map::iterator it = pages.begin(); - while(it != pages.end()){ + map::iterator it = website[0].begin(); + while(it != website[0].end()){ if(it->first == index){ write_file(target+"/index.html", it->second); }else{ - write_file(target+"/"+FileAssembler::lowercase(it->first)+".html", it->second); + write_file(target+"/"+lowercase(it->first)+".html", it->second); } it++; } - it = posts.begin(); - while(it != posts.end()){ - write_file(target+"/posts/"+FileAssembler::lowercase(it->first)+".html", it->second); + it = website[1].begin(); + while(it != website[1].end()){ + write_file(target+"/posts/"+lowercase(it->first)+".html", it->second); it++; } - if (false){ // Only copying the cached resources vs all of them - /** list resources = fa->get_cached_resources(); - - list::iterator res = resources.begin(); - while(res != resources.end()){ - error_code ec; - fs::copy(config+"resources/"+*res, target+"/"+*res,ec); - if(ec.value() != 0){ - cerr << "Error: swg: Resource file '" << *res << "' could not be copied!" << endl; - exit(6); - } - res ++; - } - **/ - }else{ - // Recursively copies all files and folders from src to target and overwrites existing files in target. - try{ - fs::copy(config+"resources/", target, fs::copy_options::overwrite_existing | fs::copy_options::recursive); - }catch (std::exception& e){ - std::cout << e.what(); // TODO better print - } + // Recursively copies all files and folders from src to target and overwrites existing files in target. + try{ + fs::copy(config+"resources/", target, fs::copy_options::overwrite_existing | fs::copy_options::recursive); + }catch (std::exception& e){ + std::cout << e.what(); // TODO better print } return 0; -- cgit v1.2.3