diff options
author | Indrajith K L | 2022-12-03 17:00:20 +0530 |
---|---|---|
committer | Indrajith K L | 2022-12-03 17:00:20 +0530 |
commit | f5c4671bfbad96bf346bd7e9a21fc4317b4959df (patch) | |
tree | 2764fc62da58f2ba8da7ed341643fc359873142f /v_windows/v/old/vlib/encoding/csv/reader.v | |
download | cli-tools-windows-f5c4671bfbad96bf346bd7e9a21fc4317b4959df.tar.gz cli-tools-windows-f5c4671bfbad96bf346bd7e9a21fc4317b4959df.tar.bz2 cli-tools-windows-f5c4671bfbad96bf346bd7e9a21fc4317b4959df.zip |
Diffstat (limited to 'v_windows/v/old/vlib/encoding/csv/reader.v')
-rw-r--r-- | v_windows/v/old/vlib/encoding/csv/reader.v | 196 |
1 files changed, 196 insertions, 0 deletions
diff --git a/v_windows/v/old/vlib/encoding/csv/reader.v b/v_windows/v/old/vlib/encoding/csv/reader.v new file mode 100644 index 0000000..dafd022 --- /dev/null +++ b/v_windows/v/old/vlib/encoding/csv/reader.v @@ -0,0 +1,196 @@ +// Copyright (c) 2019-2021 Alexander Medvednikov. All rights reserved. +// Use of this source code is governed by an MIT license +// that can be found in the LICENSE file. +module csv + +// Once interfaces are further along the idea would be to have something similar to +// go's io.reader & bufio.reader rather than reading the whole file into string, this +// would then satisfy that interface. I designed it this way to be easily adapted. +struct ErrCommentIsDelimiter { + msg string = 'encoding.csv: comment cannot be the same as delimiter' + code int +} + +struct ErrInvalidDelimiter { + msg string = 'encoding.csv: invalid delimiter' + code int +} + +struct ErrEndOfFile { + msg string = 'encoding.csv: end of file' + code int +} + +struct ErrInvalidLineEnding { + msg string = 'encoding.csv: could not find any valid line endings' + code int +} + +struct Reader { + // not used yet + // has_header bool + // headings []string + data string +pub mut: + delimiter byte + comment byte + is_mac_pre_osx_le bool + row_pos int +} + +// new_reader initializes a Reader with string data to parse +pub fn new_reader(data string) &Reader { + return &Reader{ + delimiter: `,` + comment: `#` + data: data + } +} + +// read reads a row from the CSV data. +// If successful, the result holds an array of each column's data. +pub fn (mut r Reader) read() ?[]string { + l := r.read_record() ? + return l +} + +// Once we have multi dimensional array +// pub fn (mut r Reader) read_all() ?[][]string { +// mut records := []string{} +// for { +// record := r.read_record() or { +// if err.error == err_eof.error { +// return records +// } else { +// return err +// } +// } +// records << record +// } +// return records +// } +fn (mut r Reader) read_line() ?string { + // last record + if r.row_pos == r.data.len { + return IError(&ErrEndOfFile{}) + } + le := if r.is_mac_pre_osx_le { '\r' } else { '\n' } + mut i := r.data.index_after(le, r.row_pos) + if i == -1 { + if r.row_pos == 0 { + // check for pre osx mac line endings + i = r.data.index_after('\r', r.row_pos) + if i != -1 { + r.is_mac_pre_osx_le = true + } else { + // no valid line endings found + return IError(&ErrInvalidLineEnding{}) + } + } else { + // No line ending on file + i = r.data.len - 1 + } + } + mut line := r.data[r.row_pos..i] + r.row_pos = i + 1 + // normalize win line endings (remove extra \r) + if !r.is_mac_pre_osx_le && (line.len >= 1 && line[line.len - 1] == `\r`) { + line = line[..line.len - 1] + } + return line +} + +fn (mut r Reader) read_record() ?[]string { + if r.delimiter == r.comment { + return IError(&ErrCommentIsDelimiter{}) + } + if !valid_delim(r.delimiter) { + return IError(&ErrInvalidDelimiter{}) + } + mut need_read := true + mut keep_raw := false + mut line := '' + mut fields := []string{} + mut i := -1 + for { + if need_read { + l := r.read_line() ? + if l.len <= 0 { + if keep_raw { + line += '\n' + } + continue + } else if l[0] == r.comment { + if keep_raw { + line += '\n' + l + } + continue + } else { + if keep_raw { + line += '\n' + } + line += l + } + need_read = false + keep_raw = false + } + if line.len == 0 || line[0] != `"` { // not quoted + j := line.index(r.delimiter.ascii_str()) or { + // last + fields << line[..line.len] + break + } + i = j + fields << line[..i] + line = line[i + 1..] + continue + } else { // quoted + mut need_more := true + mut has_double_quotes := false + mut j := 0 + mut n := 1 + for n < line.len { + if line[n] == `"` { + if n == line.len - 1 || line[n + 1] != `"` { + need_more = false + j = n - 1 + break + } else { + has_double_quotes = true + n++ + } + } + n++ + } + if need_more { + need_read = true + keep_raw = true + continue + } + line = line[1..] + if j + 1 == line.len { + // last record + fields << if has_double_quotes { line[..j].replace('""', '"') } else { line[..j] } + break + } + next := line[j + 1] + if next == r.delimiter { + fields << if has_double_quotes { line[..j].replace('""', '"') } else { line[..j] } + if j + 2 == line.len { + line = '' + } else { + line = line[j + 2..] + } + continue + } + } + if i <= -1 && fields.len == 0 { + return IError(&ErrInvalidDelimiter{}) + } + } + return fields +} + +fn valid_delim(b byte) bool { + return b != 0 && b != `"` && b != `\r` && b != `\n` +} |