mapleFU commented on code in PR #47524:
URL: https://github.com/apache/arrow/pull/47524#discussion_r2330390575
##########
cpp/src/arrow/csv/options.h:
##########
@@ -188,6 +188,12 @@ struct ARROW_EXPORT WriteOptions {
/// Whether to write an initial header line with column names
bool include_header = true;
+ /// \brief Quoting style of header
+ ///
+ /// If `quoting_header` is `QuotingStyle::None`, then only write quotes when
a column
+ /// name contains structural characters. Otherwise, always quote column
names.
+ QuotingStyle quoting_header = QuotingStyle::AllValid;
Review Comment:
This is a bit weird, why not follow the variables original syntax? Would
this better default as needed?
##########
cpp/src/arrow/csv/writer.cc:
##########
@@ -594,18 +604,33 @@ class CSVWriterImpl : public ipc::RecordBatchWriter {
// Only called once, as part of initialization
RETURN_NOT_OK(data_buffer_->Resize(CalculateHeaderSize(),
/*shrink_to_fit=*/false));
char* next = reinterpret_cast<char*>(data_buffer_->mutable_data());
+ int64_t no_quoting_count = 0;
for (int col = 0; col < schema_->num_fields(); ++col) {
- *next++ = '"';
- next = Escape(schema_->field(col)->name(), next);
- *next++ = '"';
+ const std::string& col_name = schema_->field(col)->name();
+ if (options_.quoting_header != QuotingStyle::None ||
Review Comment:
Personally I prefer swtich for QuotingStyle
##########
cpp/src/arrow/csv/writer.cc:
##########
@@ -594,18 +604,33 @@ class CSVWriterImpl : public ipc::RecordBatchWriter {
// Only called once, as part of initialization
RETURN_NOT_OK(data_buffer_->Resize(CalculateHeaderSize(),
/*shrink_to_fit=*/false));
char* next = reinterpret_cast<char*>(data_buffer_->mutable_data());
+ int64_t no_quoting_count = 0;
for (int col = 0; col < schema_->num_fields(); ++col) {
- *next++ = '"';
- next = Escape(schema_->field(col)->name(), next);
- *next++ = '"';
+ const std::string& col_name = schema_->field(col)->name();
+ if (options_.quoting_header != QuotingStyle::None ||
+ StopAtStructuralChar(reinterpret_cast<const
uint8_t*>(col_name.c_str()),
+ col_name.length(), options_.delimiter) !=
+ static_cast<int64_t>(col_name.length())) {
+ *next++ = '"';
+ next = Escape(schema_->field(col)->name(), next);
+ *next++ = '"';
+ } else {
+ memcpy(next, col_name.data(), col_name.size());
+ next += col_name.size();
+ no_quoting_count += 1;
+ }
if (col != schema_->num_fields() - 1) {
*next++ = options_.delimiter;
}
}
memcpy(next, options_.eol.data(), options_.eol.size());
next += options_.eol.size();
- DCHECK_EQ(reinterpret_cast<uint8_t*>(next),
+ DCHECK_EQ(reinterpret_cast<uint8_t*>(next) + no_quoting_count *
kQuoteCount,
data_buffer_->data() + data_buffer_->size());
+ if (no_quoting_count != 0) {
+ RETURN_NOT_OK(data_buffer_->Resize(
+ data_buffer_->size() - no_quoting_count * kQuoteCount,
/*shrink_to_fit=*/true));
Review Comment:
Why shrink_to_fit here?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
