More clean-up

Author: vincentlaucsb

include/internal/basic_csv_parser.cpp

@@ -248,11 +248,8 @@ namespace csv {
 #ifdef _MSC_VER
 #pragma region Specializations
 #endif
+#if !defined(__EMSCRIPTEN__)
         CSV_INLINE void MmapParser::next(size_t bytes = ITERATION_CHUNK_SIZE) {
-#if defined(__EMSCRIPTEN__)
-            (void)bytes;
-            throw std::runtime_error("MmapParser is not supported on Emscripten; use stream-based parsing.");
-#else
             // CRITICAL SECTION: Chunk Transition Logic
             // This function reads 10MB chunks and must correctly handle fields that span
             // chunk boundaries. The 'remainder' calculation below ensures partial fields
@@ -306,8 +303,8 @@ namespace csv {
             }
 
             this->mmap_pos -= (length - remainder);
-#endif
         }
+#endif
 #ifdef _MSC_VER
 #pragma endregion
 #endif

include/internal/basic_csv_parser.hpp

@@ -273,6 +273,7 @@ namespace csv {
             size_t stream_pos = 0;
         };
 
+#if !defined(__EMSCRIPTEN__)
         /** Parser for memory-mapped files
          *
          *  @par Implementation
@@ -300,5 +301,6 @@ namespace csv {
             std::string _filename;
             size_t mmap_pos = 0;
         };
+#endif
     }
 }

include/internal/csv_reader.cpp

@@ -165,12 +165,16 @@ namespace csv {
      *
      */
 	CSV_INLINE CSVReader::CSVReader(csv::string_view filename, CSVFormat format) : _format(format) {
-        auto head = internals::get_csv_head(filename);
 #if defined(__EMSCRIPTEN__)
-    using Parser = internals::StreamParser<std::ifstream>;
+    this->owned_file_stream = std::unique_ptr<std::ifstream>(new std::ifstream(std::string(filename), std::ios::binary));
+    if (!this->owned_file_stream->is_open()) {
+        throw std::runtime_error("Cannot open file " + std::string(filename));
+    }
+
+    this->init_from_stream(*this->owned_file_stream, format);
 #else
+    auto head = internals::get_csv_head(filename);
         using Parser = internals::MmapParser;
-#endif
         // Apply chunk size from format before any reading occurs
         this->_chunk_size = format.get_chunk_size();
         /** Guess delimiter and header row */
@@ -190,16 +194,9 @@ namespace csv {
         if (!format.col_names.empty())
             this->set_col_names(format.col_names);
 
-#if defined(__EMSCRIPTEN__)
-        this->owned_file_stream = std::unique_ptr<std::ifstream>(new std::ifstream(std::string(filename), std::ios::binary));
-        if (!this->owned_file_stream->is_open()) {
-            throw std::runtime_error("Cannot open file " + std::string(filename));
-        }
-        this->parser = std::unique_ptr<Parser>(new Parser(*this->owned_file_stream, format, this->col_names));
-#else
         this->parser = std::unique_ptr<Parser>(new Parser(filename, format, this->col_names)); // For C++11
-#endif
         this->initial_read();
+#endif
     }
 
     /** Return the format of the original raw CSV */

include/internal/csv_reader.hpp

@@ -183,30 +183,7 @@ namespace csv {
         template<typename TStream,
             csv::enable_if_t<std::is_base_of<std::istream, TStream>::value, int> = 0>
         CSVReader(TStream &source, CSVFormat format = CSVFormat::guess_csv()) : _format(format) {
-            auto head = internals::get_csv_head(source);
-            using Parser = internals::StreamParser<TStream>;
-
-            // Apply chunk size from format before any reading occurs
-            this->_chunk_size = format.get_chunk_size();
-
-            if (format.guess_delim()) {
-                auto guess_result = internals::_guess_format(head, format.possible_delimiters);
-                format.delimiter(guess_result.delim);
-                // Only override header if user hasn't explicitly called no_header()
-                // Note: column_names() also sets header=-1, but it populates col_names,
-                // so we can distinguish: no_header() means header=-1 && col_names.empty()
-                if (format.header != -1 || !format.col_names.empty()) {
-                    format.header = guess_result.header_row;
-                }
-                this->_format = format;
-            }
-
-            if (!format.col_names.empty())
-                this->set_col_names(format.col_names);
-
-            this->parser = std::unique_ptr<Parser>(
-                new Parser(source, format, col_names)); // For C++11
-            this->initial_read();
+            this->init_from_stream(source, format);
         }
         ///@}
 
@@ -283,8 +260,10 @@ namespace csv {
         /** Queue of parsed CSV rows */
         std::unique_ptr<RowCollection> records{new RowCollection(100)};
 
-        /** Owned file stream used for stream-based filename parsing on targets without mmap. */
+    #if defined(__EMSCRIPTEN__)
+        /** Owned file stream used by filename constructor fallback to stream parsing. */
         std::unique_ptr<std::ifstream> owned_file_stream = nullptr;
+    #endif
 
         size_t n_cols = 0;  /**< The number of columns in this CSV */
         size_t _n_rows = 0; /**< How many rows (minus header) have been read so far */
@@ -337,6 +316,36 @@ namespace csv {
             }
         }
 
+        template<typename TStream,
+            csv::enable_if_t<std::is_base_of<std::istream, TStream>::value, int> = 0>
+        void init_from_stream(TStream& source, CSVFormat format) {
+            auto head = internals::get_csv_head(source);
+            using Parser = internals::StreamParser<TStream>;
+
+            // Apply chunk size from format before any reading occurs
+            this->_chunk_size = format.get_chunk_size();
+
+            if (format.guess_delim()) {
+                auto guess_result = internals::_guess_format(head, format.possible_delimiters);
+                format.delimiter(guess_result.delim);
+                // Only override header if user hasn't explicitly called no_header()
+                // Note: column_names() also sets header=-1, but it populates col_names,
+                // so we can distinguish: no_header() means header=-1 && col_names.empty()
+                if (format.header != -1 || !format.col_names.empty()) {
+                    format.header = guess_result.header_row;
+                }
+                this->_format = format;
+            }
+
+            if (!format.col_names.empty()) {
+                this->set_col_names(format.col_names);
+            }
+
+            this->parser = std::unique_ptr<Parser>(
+                new Parser(source, format, col_names)); // For C++11
+            this->initial_read();
+        }
+
         /** Read initial chunk to get metadata */
         void initial_read() {
 #if CSV_ENABLE_THREADS