Recent changes to patches

MySQL 8 build fixes

2022-05-12T14:07:40.590000Z

cppdb is failing to build on Ubuntu systems due to MySQL8 changes. Here are two patches that Ubuntu has been carrying:
https://git.launchpad.net/ubuntu/+source/cppdb/tree/debian/patches/mysql-8.0
https://git.launchpad.net/ubuntu/+source/cppdb/tree/debian/patches/mysql8_my_bool.patch

We haven't tested if it will still build with mysql older than 8 with these patches applied.

I'm also attaching them for convenience.

minor fixes to the cppdb test.

2020-03-06T19:52:41.719000Z

minor bug fixes.
minor test connections addition.

Compile error in std::pair serialization.

2019-09-16T12:18:55.054000Z

The attached program json_pair.cpp fails to compile with lots of errors, including
/usr/local/include/cppcms/json.h:613:76: error: no matching function for call to 'cppcms::json::value::ge
t_value(const char [6]) const'
std::pair<t1,t2> pair(v.get_value<t1>("first"),v.get_value<t2>("second"));
and
/usr/local/include/cppcms/json.h:619:4: error: no matching function for call to 'cppcms::json::value::set
_value(const char [6], const int&)'
v.set_value("first",in.first);</t2></t1></t1,t2>

It can be fixed changing [g|s]et_value with [g|s]set in cppcms::json::traits< std::pair<T1,T2> >, like in the diff file.

Why does my CppCMS application not run with FastCGI configuration?

2018-09-17T15:47:00.620000Z

I follow this tutorial to create my sample application CMS00 on my Ubuntu machine. This is added to the NginX server block of /etc/nginx/sites-available/default:

location ~ ^/CMS00.*$ {
    fastcgi_pass    127.0.0.1:8080;
    fastcgi_split_path_info ^(/CMS00)((?:/.*))?$;  
    fastcgi_param   PATH_INFO       $fastcgi_path_info;  
    include         fastcgi_params;
}

Then I use this config.js file:

{  
  "service" : {  
    "api" : "fastcgi",
    "ip": "127.0.0.1",
    "port" : 8080
  }  
}

And then added the path with:

export LD_LIBRARY_PATH=/usr/local/lib

When I try to run it:

./CMS00 -c config.js

Nothing comes on. However if I change config.js to:

{  
    "service" :
    {  
        "api" : "http",  
        "port" : 8080  
    },  
    "http" :
    {  
        "script_names" : [ "/CMS00" ]  
    }  
}

It runs properly. Could someone tell me what I did wrong with the FastCGI configuration? Thank you so much.

how to load an .png file to browser using content.h of cppcms

2018-07-10T07:28:24.735000Z

Hi,

I am new to cppcms. I am trying to execute template using Linux using cppcms.
my understanding is like html data will be provided in content.h file and same can be viewed on browser.
content.h file is edited by using cppcms::widgets::text to display text
to display image what widgets need to be used
please help me on this.

I might be raising in wrong place. if I am wrong please let me know where i can get the answer.

Thanks in advance.

#21 Add bool data type to CppDB

2017-03-04T17:55:11.432000Z

Sorry! This is is missing attachment of the patch file.

Add bool data type to CppDB

2017-03-04T17:52:22.689000Z

Important! This patch works only for PostgreSQL. Applying this patch will cause all other back end drivers unable to compile.

Please check comments TODO in this diff file before applying this patch. All except PostgreSQL drivers are commented out in CMakefile.

changes made:

adding bool data type
moving PGconn* from private to public so that applications can do lower level works like this:

cppdb::postgresql::connection cppdb_pg_conn(cppdb::connection_info("CONN-STRING"));
PGconn *pg_conn=cppdb_pg_conn.pg_conn();
PQexec(pg_conn,"LISTEN notify_me_of_changes");
int sock=PQsocket(pg_conn);

This diff file comes from svn diff applied to cppdb/trunk.

#20 minor aes_encryptor.cpp cleanup

2016-11-03T07:22:09.432000Z

Regarding AES Patch...

It is exactly the reason all cryptography stuff should be reviewed under magnifier

it is wrong:

IV isn't stored it is rundom. AES CBC is self syncrhonizing algorithms so even if you don't know IV - and you don't it is rundom. Once the first block is read you have valid data. So first block is not in use.
The memset exits for very good reason - make sure that the signature does not accidentially leak. Example of an attack. If user has a bug and shows some uninitialized stuff from allocated memory to user I can send invalid cache data to server. When calculated signature is "free" it may accidentially be taken by the another malloc and exposed via "harmless" bug to user and how can can put a valid signature on session...

Regarding 2nd - good stuff.

minor aes_encryptor.cpp cleanup

2016-11-03T07:10:09.405000Z

This patch cleans up aes_encryptor.cpp and makes rpc_json.cpp more informative.

Remove unused local typedef

2015-04-14T08:53:57.832000Z

Hello,

I do this on github, but it doesn't seem followed.

https://github.com/boostorg/locale/pull/7

Regards,

Nicolas Cornu