Skip to content
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

Jump to bottom

sprintf creates wrong string #40303

Closed
mzaffalon opened this issue Apr 1, 2021 · 2 comments · Fixed by #40321
Closed

sprintf creates wrong string #40303

mzaffalon opened this issue Apr 1, 2021 · 2 comments · Fixed by #40321
Assignees
@quinnj
Labels
bug Indicates an unexpected problem or unintended behavior regression Regression in behavior compared to a previous version

Comments

@mzaffalon
Copy link
Contributor

mzaffalon commented Apr 1, 2021 •
edited
Loading

Some values are not correctly converted to string when using @sprintf:

julia> using Printf

julia> @sprintf("%+7.1f", 9.95)
"   +9.9"
julia> @sprintf("%+7.1f", 9.96) # expected "  +10.0"
"   ,0.0"

julia> versioninfo()
Julia Version 1.6.0
Commit f9720dc2eb (2021-03-24 12:55 UTC)
Platform Info:
  OS: Windows (x86_64-w64-mingw32)
  CPU: Intel(R) Xeon(R) CPU E5-1620 v3 @ 3.50GHz
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-11.0.1 (ORCJIT, haswell)
@KristofferC KristofferC added bug Indicates an unexpected problem or unintended behavior regression Regression in behavior compared to a previous version labels Apr 2, 2021
quinnj added a commit that referenced this issue Apr 2, 2021
@quinnj
Fix rounding cases for fixed-width printing in ryu routines
e0e5c5b 
Fixes #40303. When printing values to fixed widths through the
`Ryu.writefixed` or `Ryu.writeexp` routines, we have a "cleanup" section
after a value has been printed to see if it needs to be rounded given
the input precision and width. The core issue was the terminating
condition: it previously only checked if we were at the start of a
buffer or had encountered the `'-'` character. Via Printf formatting,
however, it also allows specifying the `'+'` and `' '` characters to
preceed a formatted number. Hence, in the OP, the `'1'` character was
getting "rounded" up to the `','` character. The fix here is correctly
checking if the `plus` or `space` options were passed to the routine and
if so, include those in our rounding termination check. The original
issue only reported the "plus" case for the `f` format specifier, but
the same bug affects the `e` format specifier and the "space" option.
quinnj added a commit that referenced this issue Apr 2, 2021
@quinnj
Fix rounding cases for fixed-width printing in ryu routines
fcf6779 
Fixes #40303. When printing values to fixed widths through the
`Ryu.writefixed` or `Ryu.writeexp` routines, we have a "cleanup" section
after a value has been printed to see if it needs to be rounded given
the input precision and width. The core issue was the terminating
condition: it previously only checked if we were at the start of a
buffer or had encountered the `'-'` character. Via Printf formatting,
however, it also allows specifying the `'+'` and `' '` characters to
preceed a formatted number. Hence, in the OP, the `'1'` character was
getting "rounded" up to the `','` character. The fix here is correctly
checking if the `plus` or `space` options were passed to the routine and
if so, include those in our rounding termination check. The original
issue only reported the "plus" case for the `f` format specifier, but
the same bug affects the `e` format specifier and the "space" option.
@quinnj quinnj mentioned this issue Apr 2, 2021
Merged
@quinnj
Copy link
Member

quinnj commented Apr 2, 2021

Thanks for the report @mzaffalon! A fix is up: #40321

@mzaffalon
Copy link
Contributor Author

@quinnj Thank you for the rapid fix!

KristofferC pushed a commit that referenced this issue Apr 4, 2021
@quinnj @KristofferC
Fix rounding cases for fixed-width printing in ryu routines
85d25a1 
Fixes #40303. When printing values to fixed widths through the
`Ryu.writefixed` or `Ryu.writeexp` routines, we have a "cleanup" section
after a value has been printed to see if it needs to be rounded given
the input precision and width. The core issue was the terminating
condition: it previously only checked if we were at the start of a
buffer or had encountered the `'-'` character. Via Printf formatting,
however, it also allows specifying the `'+'` and `' '` characters to
preceed a formatted number. Hence, in the OP, the `'1'` character was
getting "rounded" up to the `','` character. The fix here is correctly
checking if the `plus` or `space` options were passed to the routine and
if so, include those in our rounding termination check. The original
issue only reported the "plus" case for the `f` format specifier, but
the same bug affects the `e` format specifier and the "space" option.

(cherry picked from commit fcf6779)
KristofferC pushed a commit that referenced this issue Apr 4, 2021
@quinnj @KristofferC
Fix rounding cases for fixed-width printing in ryu routines
31e405f 
Fixes #40303. When printing values to fixed widths through the
`Ryu.writefixed` or `Ryu.writeexp` routines, we have a "cleanup" section
after a value has been printed to see if it needs to be rounded given
the input precision and width. The core issue was the terminating
condition: it previously only checked if we were at the start of a
buffer or had encountered the `'-'` character. Via Printf formatting,
however, it also allows specifying the `'+'` and `' '` characters to
preceed a formatted number. Hence, in the OP, the `'1'` character was
getting "rounded" up to the `','` character. The fix here is correctly
checking if the `plus` or `space` options were passed to the routine and
if so, include those in our rounding termination check. The original
issue only reported the "plus" case for the `f` format specifier, but
the same bug affects the `e` format specifier and the "space" option.

(cherry picked from commit fcf6779)
KristofferC pushed a commit that referenced this issue Apr 4, 2021
@quinnj @KristofferC
Fix rounding cases for fixed-width printing in ryu routines
bef3e1a 
Fixes #40303. When printing values to fixed widths through the
`Ryu.writefixed` or `Ryu.writeexp` routines, we have a "cleanup" section
after a value has been printed to see if it needs to be rounded given
the input precision and width. The core issue was the terminating
condition: it previously only checked if we were at the start of a
buffer or had encountered the `'-'` character. Via Printf formatting,
however, it also allows specifying the `'+'` and `' '` characters to
preceed a formatted number. Hence, in the OP, the `'1'` character was
getting "rounded" up to the `','` character. The fix here is correctly
checking if the `plus` or `space` options were passed to the routine and
if so, include those in our rounding termination check. The original
issue only reported the "plus" case for the `f` format specifier, but
the same bug affects the `e` format specifier and the "space" option.

(cherry picked from commit fcf6779)
ElOceanografo pushed a commit to ElOceanografo/julia that referenced this issue May 4, 2021
@quinnj @ElOceanografo
Fix rounding cases for fixed-width printing in ryu routines
e6e596b 
Fixes JuliaLang#40303. When printing values to fixed widths through the
`Ryu.writefixed` or `Ryu.writeexp` routines, we have a "cleanup" section
after a value has been printed to see if it needs to be rounded given
the input precision and width. The core issue was the terminating
condition: it previously only checked if we were at the start of a
buffer or had encountered the `'-'` character. Via Printf formatting,
however, it also allows specifying the `'+'` and `' '` characters to
preceed a formatted number. Hence, in the OP, the `'1'` character was
getting "rounded" up to the `','` character. The fix here is correctly
checking if the `plus` or `space` options were passed to the routine and
if so, include those in our rounding termination check. The original
issue only reported the "plus" case for the `f` format specifier, but
the same bug affects the `e` format specifier and the "space" option.
antoine-levitt pushed a commit to antoine-levitt/julia that referenced this issue May 9, 2021
@quinnj @antoine-levitt
Fix rounding cases for fixed-width printing in ryu routines
afcebf0 
Fixes JuliaLang#40303. When printing values to fixed widths through the
`Ryu.writefixed` or `Ryu.writeexp` routines, we have a "cleanup" section
after a value has been printed to see if it needs to be rounded given
the input precision and width. The core issue was the terminating
condition: it previously only checked if we were at the start of a
buffer or had encountered the `'-'` character. Via Printf formatting,
however, it also allows specifying the `'+'` and `' '` characters to
preceed a formatted number. Hence, in the OP, the `'1'` character was
getting "rounded" up to the `','` character. The fix here is correctly
checking if the `plus` or `space` options were passed to the routine and
if so, include those in our rounding termination check. The original
issue only reported the "plus" case for the `f` format specifier, but
the same bug affects the `e` format specifier and the "space" option.
johanmon pushed a commit to johanmon/julia that referenced this issue Jul 5, 2021
@quinnj @johanmon
Fix rounding cases for fixed-width printing in ryu routines
d6669b4 
Fixes JuliaLang#40303. When printing values to fixed widths through the
`Ryu.writefixed` or `Ryu.writeexp` routines, we have a "cleanup" section
after a value has been printed to see if it needs to be rounded given
the input precision and width. The core issue was the terminating
condition: it previously only checked if we were at the start of a
buffer or had encountered the `'-'` character. Via Printf formatting,
however, it also allows specifying the `'+'` and `' '` characters to
preceed a formatted number. Hence, in the OP, the `'1'` character was
getting "rounded" up to the `','` character. The fix here is correctly
checking if the `plus` or `space` options were passed to the routine and
if so, include those in our rounding termination check. The original
issue only reported the "plus" case for the `f` format specifier, but
the same bug affects the `e` format specifier and the "space" option.
staticfloat pushed a commit that referenced this issue Dec 23, 2022
@quinnj @staticfloat
Fix rounding cases for fixed-width printing in ryu routines
2524f13 
Fixes #40303. When printing values to fixed widths through the
`Ryu.writefixed` or `Ryu.writeexp` routines, we have a "cleanup" section
after a value has been printed to see if it needs to be rounded given
the input precision and width. The core issue was the terminating
condition: it previously only checked if we were at the start of a
buffer or had encountered the `'-'` character. Via Printf formatting,
however, it also allows specifying the `'+'` and `' '` characters to
preceed a formatted number. Hence, in the OP, the `'1'` character was
getting "rounded" up to the `','` character. The fix here is correctly
checking if the `plus` or `space` options were passed to the routine and
if so, include those in our rounding termination check. The original
issue only reported the "plus" case for the `f` format specifier, but
the same bug affects the `e` format specifier and the "space" option.

(cherry picked from commit fcf6779)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@quinnj quinnj

Labels
bug Indicates an unexpected problem or unintended behavior regression Regression in behavior compared to a previous version
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

Fix rounding cases for fixed-width printing in ryu routines JuliaLang/julia
3 participants
@KristofferC @mzaffalon @quinnj