From c858f8a9146f20525152bacf1913415ff7d44acc Mon Sep 17 00:00:00 2001
From: Alex Macocian <amacocian@yahoo.com>
Date: Fri, 17 Nov 2023 13:37:08 +0100
Subject: [PATCH] Implemented complete parsing

---
 source/gMod_FileLoader.cpp | 126 +++++++++++++++++++++++++++++++++++--
 1 file changed, 121 insertions(+), 5 deletions(-)

diff --git a/source/gMod_FileLoader.cpp b/source/gMod_FileLoader.cpp
index 52595f0..2c44d35 100644
--- a/source/gMod_FileLoader.cpp
+++ b/source/gMod_FileLoader.cpp
@@ -38,7 +38,7 @@ std::vector<TpfEntry> gMod_FileLoader::GetTpfContents()
         auto zipArchive = libzippp::ZipArchive::fromBuffer(buffer.data(), buffer.size(), false, password);
         zipArchive->setErrorHandlerCallback([](const std::string& message, const std::string& strerror, int zip_error_code, int system_error_code) -> void {
             Message("GetTpfContents: %s %s %d %d\n", message.c_str(), strerror.c_str(), zip_error_code, system_error_code);
-        });
+            });
         zipArchive->open();
         LoadEntries(*zipArchive, entries);
     }
@@ -65,14 +65,130 @@ std::vector<TpfEntry> gMod_FileLoader::GetFileContents()
     return entries;
 }
 
-void gMod_FileLoader::LoadEntries(libzippp::ZipArchive& archive, std::vector<TpfEntry>& entries)
-{
-    // Iterate over the files in the zip archive
+void ParseSimpleArchive(libzippp::ZipArchive& archive, std::vector<TpfEntry>& entries) {
     for (const auto& entry : archive.getEntries()) {
         if (entry.isFile()) {
+            //TODO: #6 - Implement regex search
             const auto dataPtr = entry.readAsBinary();
             const auto size = entry.getSize();
-            entries.push_back({entry.getName(), entry.getName(), 0, dataPtr, size}); // Assuming ZipEntry is a string
+            auto name = entry.getName();
+
+            // Remove the part before the last underscore (if any)
+            size_t firstIndex = name.find_last_of('_');
+            while (firstIndex != std::string::npos) {
+                if (firstIndex >= entry.getName().length() - 1) {
+                    name = entry.getName();
+                }
+                else {
+                    name = entry.getName().substr(firstIndex + 1);
+                }
+
+                firstIndex = name.find_last_of('_');
+            }
+
+            // Remove the file extension (if any)
+            size_t lastIndex = name.find_last_of('.');
+            if (lastIndex != std::string::npos) {
+                name = name.substr(0, lastIndex);
+            }
+
+            uint32_t crcHash;
+            try {
+                crcHash = std::stoul(name, nullptr, 16);
+            }
+            catch (const std::invalid_argument& e) {
+                Message("Failed to parse %s as a hash", name.c_str());
+                continue;
+            }
+            catch (const std::out_of_range& e) {
+                Message("Out of range while parsing %s as a hash", name.c_str());
+                continue;
+            }
+
+            entries.push_back({ name, entry.getName(), crcHash, dataPtr, size });
         }
     }
 }
+
+void ParseTexmodArchive(std::vector<std::string>& lines, libzippp::ZipArchive& archive, std::vector<TpfEntry>& entries) {
+    for (const auto& line : lines) {
+        std::istringstream iss(line);
+        std::string part;
+        std::vector<std::string> splits;
+
+        // Split the line by '|'
+        while (std::getline(iss, part, '|')) {
+            splits.push_back(part);
+        }
+
+        if (splits.size() != 2) {
+            continue;
+        }
+
+        std::string addrstr = splits[0];
+        std::string path = splits[1];
+
+        // Remove unwanted characters from the beginning of the path
+        while (!path.empty() && (path[0] == '.' && (path[1] == '/' || path[1] == '\\')) || path[0] == '/' || path[0] == '\\') {
+            path.erase(0, 1);
+        }
+
+        // Remove trailing newline and carriage return characters
+        size_t endpos = path.find_last_not_of("\r\n");
+        if (endpos != std::string::npos) {
+            path.erase(endpos + 1);
+        }
+        else if (!path.empty()) {
+            path.clear();
+        }
+
+        const auto entry = archive.getEntry(path);
+        if (entry.isNull()) {
+            continue;
+        }
+
+        if (!entry.isFile()) {
+            continue;
+        }
+
+        const auto dataPtr = entry.readAsBinary();
+        const auto size = entry.getSize();
+        uint32_t crcHash;
+        try {
+            crcHash = std::stoul(addrstr, nullptr, 16);
+        }
+        catch (const std::invalid_argument& e) {
+            Message("Failed to parse %s as a hash", addrstr.c_str());
+            continue;
+        }
+        catch (const std::out_of_range& e) {
+            Message("Out of range while parsing %s as a hash", addrstr.c_str());
+            continue;
+        }
+
+        entries.push_back({ addrstr, entry.getName(), crcHash, dataPtr, size });
+    }
+}
+
+void gMod_FileLoader::LoadEntries(libzippp::ZipArchive& archive, std::vector<TpfEntry>& entries)
+{
+    // Iterate over the files in the zip archive
+    auto maybeDefiniton = archive.getEntry("texmod.def");
+    if (maybeDefiniton.isNull() || !maybeDefiniton.isFile()) {
+        ParseSimpleArchive(archive, entries);
+    }
+    else {
+        const auto def = maybeDefiniton.readAsText();
+        std::istringstream iss(def);
+        std::vector<std::string> lines;
+        std::string line;
+
+        while (std::getline(iss, line)) {
+            lines.push_back(line);
+        }
+
+        ParseTexmodArchive(lines, archive, entries);
+    }
+}
+
+