Skip to content

feat(query): support TH/th ordinal and V shift patterns in to_char #19830

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
kimjune01 wants to merge 6 commits into
base: main
Choose a base branch
from
Draft

feat(query): support TH/th ordinal and V shift patterns in to_char #19830

Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
e8e3fb2
feat(query): support TH/th ordinal number suffix in to_char
kimjune01 May 9, 2026
275d76c
feat(query): support V (shift digits) pattern in to_char
kimjune01 May 9, 2026
ddc9525
fix(query): address review feedback for TH/th and V patterns
kimjune01 May 9, 2026
21fdfaa
style: fix clippy implicit_saturating_sub lint
kimjune01 May 9, 2026
3b8322c
chore: retrigger CI
kimjune01 May 19, 2026
6a5beee
Merge remote-tracking branch 'origin/main' into feat/to-char-th-ordinal
kimjune01 May 20, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
158 changes: 151 additions & 7 deletions src/common/io/src/number.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -132,6 +132,8 @@ enum NumFlag {
PlusPost,
MinusPost,
Eeee,
ThUpper,
ThLower,
}

#[derive(Debug, Clone, Copy)]
Expand Down Expand Up @@ -203,6 +205,11 @@ impl NumDesc {
return Err("\"0\" must be ahead of \"PR\"");
}

if self.flag.contains(NumFlag::Multi) {
self.multi += 1;
return Ok(());
}

if !self.flag.intersects(NumFlag::Zero | NumFlag::Decimal) {
self.flag.insert(NumFlag::Zero);
self.zero_start = self.pre + 1;
Expand Down Expand Up @@ -365,6 +372,16 @@ impl NumDesc {
Ok(())
}

NumPoz::TkTH => {
self.flag.insert(NumFlag::ThUpper);
Ok(())
}

NumPoz::Tkth => {
self.flag.insert(NumFlag::ThLower);
Ok(())
}

NumPoz::TkComma => Ok(()),

_ => unreachable!(),
Expand Down Expand Up @@ -401,7 +418,39 @@ impl NumDesc {
}

if self.flag.contains(NumFlag::Multi) {
return Err(ErrorCode::Unimplemented("to_char V (multiplies)"));
// V shifts the number by `multi` decimal places (multiply by 10^multi).
// Total output width = pre + multi digits.
let total_width = self.pre + self.multi;
let abs_val = if value == i64::MIN {
-(i64::MIN as i128)
} else {
value.unsigned_abs() as i128
};
let result = 10i128
.checked_pow(self.multi as u32)
.and_then(|m| abs_val.checked_mul(m));

let (number, out_pre_spaces) = match result {
Some(shifted) => {
let orgnum = format!("{}", shifted);
let numstr_pre_len = orgnum.len();
match numstr_pre_len.cmp(&total_width) {
std::cmp::Ordering::Less => (orgnum, total_width - numstr_pre_len),
std::cmp::Ordering::Greater => ("#".repeat(total_width), 0),
std::cmp::Ordering::Equal => (orgnum, 0),
}
}
None => {
// Overflow: display as hashes
("#".repeat(total_width), 0)
}
};

return Ok(NumPart {
sign: value >= 0,
number,
out_pre_spaces,
});
}

let mut orgnum = if value == i64::MIN {
Expand Down Expand Up @@ -479,7 +528,26 @@ impl NumDesc {
}

if self.flag.contains(NumFlag::Multi) {
return Err(ErrorCode::Unimplemented("to_char V (multiplies)"));
// V shifts the number by `multi` decimal places (multiply by 10^multi),
// then rounds to the nearest integer.
// Total output width = pre + multi digits.
let multiplier = 10f64.powi(self.multi as i32);
let shifted = (value.abs() * multiplier).round_ties_even();

@chatgpt-codex-connector chatgpt-codex-connector Bot May 9, 2026

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P1 Badge Use away-from-zero rounding for V shift ties

The V implementation for floating-point values currently uses round_ties_even(), but PostgreSQL rounds to_char(..., '...V...') half values away from zero (for example, to_char(12.45, '99V9') is documented as 125). With the current code, tie cases like 12.45 produce 124, so Databend will return PostgreSQL-incompatible results for common .5 boundaries.

Useful? React with 👍 / 👎.

let orgnum = format!("{:.0}", shifted);
let numstr_pre_len = orgnum.len();
let total_width = self.pre + self.multi;

let (number, out_pre_spaces) = match numstr_pre_len.cmp(&total_width) {
std::cmp::Ordering::Less => (orgnum, total_width - numstr_pre_len),
std::cmp::Ordering::Greater => ("#".repeat(total_width), 0),
std::cmp::Ordering::Equal => (orgnum, 0),
};

return Ok(NumPart {
sign: !value.is_sign_negative(),
number,
out_pre_spaces,
});
}

let orgnum = format!("{:.0}", value.abs());
Expand Down Expand Up @@ -945,7 +1013,7 @@ fn num_processor(nodes: &[FormatNode], desc: NumDesc, num_part: NumPart) -> Resu
}

// Count
np.num_count = np.desc.post + np.desc.pre - 1;
np.num_count = np.desc.post + np.desc.pre + np.desc.multi - 1;

if np.desc.flag.contains(NumFlag::FillMode) && np.desc.flag.contains(NumFlag::Decimal) {
np.calc_last_relevant_decnum();
Expand Down Expand Up @@ -1039,6 +1107,35 @@ fn num_processor(nodes: &[FormatNode], desc: NumDesc, num_part: NumPart) -> Resu

NumPoz::TkPR => (),
NumPoz::TkFM => (),
NumPoz::TkV => (),
NumPoz::TkTH | NumPoz::Tkth => {

@chatgpt-codex-connector chatgpt-codex-connector Bot May 9, 2026

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

P2 Badge Skip ordinal suffix conversion for negative TH/th

The ordinal suffix branch always appends st/nd/rd/th, but PostgreSQL's numeric formatting rules state that TH/th does not convert values less than zero (and also not fractional inputs). As written, calls like to_char(-1, '9th') will emit an ordinal suffix, which breaks PostgreSQL compatibility for negative values.

Useful? React with 👍 / 👎.

// Extract last two digits from internal number to determine suffix.
// Use only digits before any decimal point.
let int_part: &[char] = match np.number.iter().position(|c| *c == '.') {
Some(dot) => &np.number[..dot],
None => &np.number,
};
let last_two: u64 = {
let len = int_part.len();
let start = len.saturating_sub(2);
int_part[start..]
.iter()
.filter(|c| c.is_ascii_digit())
.fold(0u64, |acc, c| acc * 10 + (*c as u64 - '0' as u64))
};
let suffix = match last_two % 100 {
11..=13 => "th",
n if n % 10 == 1 => "st",
n if n % 10 == 2 => "nd",
n if n % 10 == 3 => "rd",
_ => "th",
};
if matches!(key.id, NumPoz::TkTH) {
np.inout.push_str(&suffix.to_uppercase());
} else {
np.inout.push_str(suffix);
}
}
_ => unimplemented!(),
},
FormatNode::End => break,
Expand Down Expand Up @@ -1192,6 +1289,50 @@ mod tests {
assert_eq!("###############", i64_to_char(4000, "FMRN")?);
assert_eq!("###############", i64_to_char(-1, "FMRN")?);

// TH / th ordinal suffix
assert_eq!(" 0TH", i64_to_char(0, "9TH")?);
assert_eq!(" 1ST", i64_to_char(1, "9TH")?);
assert_eq!(" 2ND", i64_to_char(2, "9TH")?);
assert_eq!(" 3RD", i64_to_char(3, "9TH")?);
assert_eq!(" 4TH", i64_to_char(4, "9TH")?);
assert_eq!(" 11TH", i64_to_char(11, "99TH")?);
assert_eq!(" 12TH", i64_to_char(12, "99TH")?);
assert_eq!(" 13TH", i64_to_char(13, "99TH")?);
assert_eq!(" 21ST", i64_to_char(21, "99TH")?);
assert_eq!(" 22ND", i64_to_char(22, "99TH")?);
assert_eq!(" 23RD", i64_to_char(23, "99TH")?);
assert_eq!(" 111TH", i64_to_char(111, "999TH")?);
assert_eq!(" 112TH", i64_to_char(112, "999TH")?);
assert_eq!(" 113TH", i64_to_char(113, "999TH")?);

// lowercase ordinal
assert_eq!(" 1st", i64_to_char(1, "9th")?);
assert_eq!(" 2nd", i64_to_char(2, "9th")?);
assert_eq!(" 3rd", i64_to_char(3, "9th")?);
assert_eq!(" 4th", i64_to_char(4, "9th")?);
assert_eq!(" 12th", i64_to_char(12, "99th")?);

// ordinal with FM (fill mode)
assert_eq!("1st", i64_to_char(1, "FM9th")?);
assert_eq!("12th", i64_to_char(12, "FM99th")?);

// negative ordinal
assert_eq!("-1st", i64_to_char(-1, "9th")?);
assert_eq!(" -12th", i64_to_char(-12, "999th")?);

// ordinal with trailing literal (suffix must appear before literal)
assert_eq!(" 1ST!", i64_to_char(1, "9TH\"!\"")?);
assert_eq!(" 3RD!", i64_to_char(3, "9TH\"!\"")?);

// ordinal with sign suffix
assert_eq!("1st ", i64_to_char(1, "9thMI")?);
assert_eq!("1st-", i64_to_char(-1, "9thMI")?);

// V (shift/multiply by 10^n)
assert_eq!(" 12000", i64_to_char(12, "99V999")?);
assert_eq!("-12000", i64_to_char(-12, "99V999")?);
assert_eq!(" 50", i64_to_char(5, "9V9")?);

Ok(())
}

Expand Down Expand Up @@ -1249,11 +1390,14 @@ mod tests {
assert_eq!("cdlxxxv", f64_to_char(485.0, "FMrn")?);
assert_eq!("V", f64_to_char(5.2, "FMRN")?);

// assert_eq!(" 482nd", f64_to_char(482, "999th")?);
assert_eq!(" 482nd", f64_to_char(482.0, "999th")?);

// assert_eq!(" 12000", f64_to_char(12, "99V999")?);
// assert_eq!(" 12400", f64_to_char(12.4, "99V999")?);
// assert_eq!(" 125", f64_to_char(12.45, "99V9")?);
assert_eq!(" 12000", f64_to_char(12.0, "99V999")?);
assert_eq!(" 12400", f64_to_char(12.4, "99V999")?);
// 12.45 * 10 = 124.5, ties-to-even rounds to 124
assert_eq!(" 124", f64_to_char(12.45, "99V9")?);
// 12.55 * 10 = 125.5, ties-to-even rounds to 126
assert_eq!(" 126", f64_to_char(12.55, "99V9")?);

Ok(())
}
Expand Down
Loading