You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have usernames in Redmine containing Danish ø (HTML ) and å (HTML ) which ends up being wrongly formatted on the Munin graphs.
Output from MySQL and hence the plugin seems to be correct utf-8, they are interpreted fine by my shell but munin-graph doesn't render them correctly.
utf-8 char
html entity
hex
Munin png representation
ø
ø (ø)
c3 b8
å
å (å)
c3 a5
According to the Munin docs (wiki) any chars except # and \ should be valid but no expected charset is specified. I have tried to dig into the perl code to see where and how the png are generated but haven't got to the bottom of it.
The rendered chars seems to be consistent with the 8859-1 codepage/charset. However, I do not get the expected result when changing the encoding from the current utf-8 to iso-8859-1 in line 108-109 as below. Now instead of the expected 0xf8 for ø in 8859-1 a value of 0x72 is output, for å (0xe5 in 8859-1) a value of 0x73.
Sorry, I misread the raw output from the plugin and label updating on the website graphs seems to be delayed some cycles.
The above patch correctly encodes the usernames. Tested using Munin 2.0.25 from Ubuntu 16.04 LTS. I have committed the change to a fork at https://github.com/VestergaardCompany/redmine_stats and requested a pull in #3.
I have usernames in Redmine containing Danish ø (HTML ) and å (HTML ) which ends up being wrongly formatted on the Munin graphs.
Output from MySQL and hence the plugin seems to be correct utf-8, they are interpreted fine by my shell but munin-graph doesn't render them correctly.
ø
(ø)å
(å)According to the Munin docs (wiki) any chars except # and \ should be valid but no expected charset is specified. I have tried to dig into the perl code to see where and how the png are generated but haven't got to the bottom of it.
The rendered chars seems to be consistent with the 8859-1 codepage/charset. However, I do not get the expected result when changing the encoding from the current utf-8 to iso-8859-1 in line 108-109 as below. Now instead of the expected 0xf8 for ø in 8859-1 a value of 0x72 is output, for å (0xe5 in 8859-1) a value of 0x73.
The text was updated successfully, but these errors were encountered: