Ada Mode


Next: , Previous: (dir), Up: (dir)


Next: , Previous: Top, Up: Top

1 Overview

The Emacs mode for programming in Ada helps the user in understanding existing code and facilitates writing new code.

When the Gnu Ada compiler GNAT is used, the cross-reference information output by the compiler is used to provide powerful code navigation (jump to definition, find all uses, etc).

When you open a file with a file extension of .ads or .adb, Emacs will automatically load and activate Ada mode.

Ada mode works without any customization, if you are using the GNAT compiler (https://libre2.adacore.com/) and the GNAT default naming convention.

You must customize a few things if you are using a different compiler or file naming convention; See Other compiler, See Non-standard file names.

In addition, you may want to customize the indentation, capitalization, and other things; See Other customization.

Finally, for large Ada projects, you will want to set up an Emacs Ada mode project file for each project; See Project files. Note that these are different from the GNAT project files used by gnatmake and other GNAT commands.

See the Emacs info manual, section 'Running Debuggers Under Emacs', for general information on debugging.


Next: , Previous: Overview, Up: Top

2 Installation

Ada mode is part of the standard Emacs distribution; if you use that, no files need to be installed.

Ada mode is also available as a separate distribution, from the Emacs Ada mode website http://stephe-leake.org/emacs/ada-mode/emacs-ada-mode.html. The separate distribution may be more recent.

For installing the separate distribution, see the README file in the distribution.

To see what version of Ada mode you have installed, do <M-x ada-mode-version>.

The following files are provided with the Ada mode distribution:


Next: , Previous: Installation, Up: Top

3 Customizing Ada mode

Here we assume you are familiar with setting variables in Emacs, either thru 'customize' or in elisp (in your .emacs file). For a basic introduction to customize, elisp, and Emacs in general, see the tutorial in The GNU Emacs Manual.

These global Emacs settings are strongly recommended (put them in your .emacs):

     (global-font-lock-mode t)
     (transient-mark-mode t)

`(global-font-lock-mode t)' turns on syntax highlighting for all buffers (it is off by default because it may be too slow for some machines).

`(transient-mark-mode t)' highlights selected text.

See the Emacs help for each of these variables for more information.


Next: , Previous: Customization, Up: Customization

3.1 Non-standard file names

By default, Ada mode is configured to use the GNAT file naming convention, where file names are a simple modification of the Ada names, and the extension for specs and bodies are `.ads' and `.adb', respectively.

Ada mode uses the file extentions to allow moving from a package body to the corresponding spec and back.

Ada mode supports a list of alternative file extensions for specs and bodies.

For instance, if your spec and bodies files are called unit_s.ada and unit_b.ada, respectively, you can add the following to your .emacs file:

     (ada-add-extensions "_s.ada" "_b.ada")

You can define additional extensions:

     (ada-add-extensions ".ads" "_b.ada")
     (ada-add-extensions ".ads" ".body")

This means that whenever Ada mode looks for the body for a file whose extension is .ads, it will take the first available file that ends with either .adb, _b.ada or .body.

Simililarly, if Ada mode is looking for a spec, it will look for .ads or _s.ada.

If the filename is not derived from the Ada name following the GNAT convention, things are a little more complicated. You then need to rewrite the function ada-make-filename-from-adaname. Doing that is beyond the scope of this manual; see the current definitions in ada-mode.el and ada-xref.el for examples.


Next: , Previous: Non-standard file names, Up: Customization

3.2 Other compiler

By default, Ada mode is configured to use the Gnu Ada compiler GNAT.

To use a different Ada compiler, you must specify the command lines used to run that compiler, either in lisp variables or in Emacs Ada mode project files. See Project file variables for the list of project variables, and the corresponding lisp variables.


Previous: Other compiler, Up: Customization

3.3 Other customization

All user-settable Ada mode variables can be set via the menu `Ada | Customize'. Click on the `Help' button there for help on using customize.

To modify a specific variable, you can directly call the function customize-variable; just type M-x customize-variable <RET> variable-name <RET>).

Alternately, you can specify variable settings in the Emacs configuration file, .emacs. This file is coded in Emacs lisp, and the syntax to set a variable is the following:

     (setq variable-name value)


Next: , Previous: Customization, Up: Top

4 Compiling Executing

Ada projects can be compiled, linked, and executed using commands on the Ada menu. All of these commands can be customized via a project file (see Project files), but the defaults are sufficient for using the GNAT compiler for simple projects (single files, or several files in a single directory).

Even when no project file is used, the GUI project editor (menu <Ada | Project | Edit>) shows the settings of the various project file variables referenced here.


Next: , Previous: Compiling Executing, Up: Compiling Executing

4.1 Compile commands

Here are the commands for building and using an Ada project, as listed in the Ada menu.

In multi-file projects, there must be one file that is the main program. That is given by the main project file variable; it defaults to the current file if not yet set, but is also set by the “set main and build” command.

Check file
Compiles the current file in syntax check mode, by running check_cmd defined in the current project file. This typically runs faster than full compile mode, speeding up finding and fixing compilation errors.

This sets main only if it has not been set yet.

Compile file
Compiles the current file, by running comp_cmd from the current project file.

This does not set main.

Set main and Build
Sets main to the current file, then executes the Build command.
Show main
Display main in the message buffer.
Build
Compiles all obsolete units of the current main, and links main, by running make_cmd from the current project.

This sets main only if it has not been set yet.

Run
Executes the main program in a shell, displayed in a separate Emacs buffer. This runs run_cmd from the current project. The execution buffer allows for interactive input/output.

To modify the run command, in particular to provide or change the command line arguments, type <C-u> before invoking the command.

This command is not available for a cross-compilation toolchain.

It is important when using these commands to understand how main is used and changed.

Build runs 'gnatmake' on the main unit. During a typical edit/compile session, this is the only command you need to invoke, which is why it is bound to <C-c C-c>. It will compile all files needed by the main unit, and display compilation errors in any of them.

Note that Build can be invoked from any Ada buffer; typically you will be fixing errors in files other than the main, but you don't have to switch back to the main to invoke the compiler again.

Novices and students typically work on single-file Ada projects. In this case, <C-c C-m> will normally be the only command needed; it will build the current file, rather than the last-built main.

There are three ways to change main:

  1. Invoke <Ada | Set main and Build>, which sets main to the current file.
  2. Invoke <Ada | Project | Edit>, edit main and main, and click <[save]>
  3. Invoke <Ada | Project | Load>, and load a project file that specifies main


Previous: Compile commands, Up: Compiling Executing

4.2 Compiler errors

The Check file, Compile file, and Build commands all place compilation errors in a separate buffer named *compilation*.

Each line in this buffer will become active: you can simply click on it with the middle button of the mouse, or move point to it and press <RET>. Emacs will then display the relevant source file and put point on the line and column where the error was found.

You can also press the C-x ` key (next-error), and Emacs will jump to the first error. If you press that key again, it will move you to the second error, and so on.

Some error messages might also include references to other files. These references are also clickable in the same way, or put point after the line number and press <RET>.


Next: , Previous: Compiling Executing, Up: Top

5 Project files

An Emacs Ada mode project file specifies what directories hold sources for your project, and allows you to customize the compilation commands and other things on a per-project basis.

Note that Ada mode project files `*.adp' are different than GNAT compiler project files `*.gpr'. However, Emacs Ada mode can use a GNAT project project file to specify the project directories. If no other customization is needed, a GNAT project file can be used without an Emacs Ada mode project file.


Next: , Previous: Project files, Up: Project files

5.1 Project File Overview

Project files have a simple syntax; they may be edited directly. Each line specifies a project variable name and its value, separated by “=”:

     src_dir=/Projects/my_project/src_1
     src_dir=/Projects/my_project/src_2

Some variables (like src_dir) are lists; multiple occurances are concatenated.

There must be no space between the variable name and “=”, and no trailing spaces.

Alternately, a GUI editor for project files is available (see GUI Editor). It uses Emacs widgets, similar to Emacs customize.

The GUI editor also provides a convenient way to view current project settings, if they have been modified using menu commands rather than by editing the project file.

After the first Ada mode build command is invoked, there is always a current project file, given by the lisp variable ada-prj-default-project-file. Currently, the only way to show the current project file is to invoke the GUI editor.

To find the project file the first time, Ada mode uses the following search algorithm:

This algorithm always sets ada-prj-default-project-file, even when the file does not actually exist.

To change the project file before or after the first one is found, invoke <Ada | Project | Load ...>.

Or, in lisp, evaluate (ada-set-default-project-file "/path/file.adp"). This sets ada-prj-default-project-file, and reads the project file.

You can also specify a GNAT project file to <Ada | Project | Load ...> or ada-set-default-project-file. Emacs Ada mode checks the file extension; if it is .gpr, the file is treated as a GNAT project file. Any other extension is treated as an Emacs Ada mode project file.


Next: , Previous: Project File Overview, Up: Project files

5.2 GUI Editor

The project file editor is invoked with the menu `Ada | Projects | Edit'.

Once in the buffer for editing the project file, you can save your modification using the `[save]' button at the bottom of the buffer, or the C-x C-s binding. To cancel your modifications, kill the buffer or click on the `[cancel]' button.


Previous: GUI Editor, Up: Project files

5.3 Project file variables

The following variables can be defined in a project file; some can also be defined in lisp variables.

To set a project variable that is a list, specify each element of the list on a separate line in the project file.

Any project variable can be referenced in other project variables, using a shell-like notation. For instance, if the variable comp_cmd contains ${comp_opt}, the value of the comp_opt variable will be substituted when comp_cmd is used.

In addition, process environment variables can be referenced using the same syntax, or the normal $var syntax.

Most project variables have defaults that can be changed by setting lisp variables; the table below identifies the lisp variable for each project variable. Lisp variables corresponding to project variables that are lists are lisp lists.

In general, project variables are evaluated when referenced in Emacs Ada mode commands. Relative file paths are expanded to absolute relative to ${build_dir}.

Here is the list of variables. In the default values, the current directory "." is the project file directory.

ada_project_path_sep [default: ":" or ";"]
Path separator for ADA_PROJECT_PATH. It defaults to the correct value for a native implementation of GNAT for the current operating system. The user must override this when using Windows native GNAT with Cygwin Emacs, and perhaps in other cases.

Lisp variable: ada-prj-ada-project-path-sep.

ada_project_path [default: ""]
A list of directories to search for GNAT project files.

If set, the ADA_PROJECT_PATH process environment variable is set to this value in the Emacs process when the Emacs Ada mode project is selected via menu `Ada | Project | Load'.

For ada_project_path, relative file paths are expanded to absolute when the Emacs Ada project file is read, rather than when the project file is selected.

For example if the project file is in the directory /home/myproject, the environment variable GDS_ROOT is set to /home/shared, and the project file contains:

          ada_project_path_sep=:
          ada_project_path=$GDS_ROOT/makerules
          ada_project_path=../opentoken
     

the environment variable ADA_PROJECT_PATH will be set to "/home/shared/makerules:/home/opentoken/".

The default value is not the current value of this environment variable, because that will typically have been set by another project, and will therefore be incorrect for this project.

If you have the environment variable set correctly for all of your projects, you do not need to set this project variable.

bind_opt [default: ""]
Holds user binder options; used in the default build commands.

Lisp variable: ada-prj-default-bind-opt.

build_dir [default: "."]
The compile commands will be issued in this directory.
casing [default: ("~/.emacs_case_exceptions")
List of files containing casing exceptions. See the help on ada-case-exception-file for more info.

Lisp variable: ada-case-exception-file.

check_cmd [default: "${cross_prefix}gnatmake -u -c -gnatc ${gnatmake_opt} ${full_current} -cargs ${comp_opt}"]
Command used to syntax check a single file. The name of the file is substituted for full_current.

Lisp variable: ada-prj-default-check-cmd

comp_cmd [default: "${cross_prefix}gnatmake -u -c ${gnatmake_opt} ${full_current} -cargs ${comp_opt}"]
Command used to compile a single file. The name of the file is substituted for full_current.

Lisp variable: ada-prj-default-comp-cmd.

comp_opt [default: "-gnatq -gnatQ"]
Holds user compiler options; used in the default compile commands. The default value tells gnatmake to generate library files for cross-referencing even when there are errors.

If source code for the project is in multiple directories, the appropriate compiler options must be added here. Set source search path for examples of this. Alternately, GNAT project files may be used; Use GNAT project file.

Lisp variable: ada-prj-default-comp-opt.

cross_prefix [default: ""]
Name of target machine in a cross-compilation environment. Used in default compile and build commands.
debug_cmd [default: "${cross_prefix}gdb ${main}"]
Command used to debug the application

Lisp variable: ada-prj-default-debugger.

debug_post_cmd [default: ""]
Command executed after debug_cmd.
debug_pre_cmd [default: "cd ${build_dir}"]
Command executed before debug_cmd.
gnatfind_opt [default: "-rf"]
Holds user gnatfind options; used in the default find commands.

Lisp variable: ada-prj-gnatfind-switches.

gnatmake_opt [default: "-g"]
Holds user gnatmake options; used in the default build commands.

Lisp variable: ada-prj-default-gnatmake-opt.

gpr_file [default: ""]
Specify GNAT project file.

If set, the source and object directories specified in the GNAT project file are appended to src_dir and obj_dir. This allows specifying Ada source directories with a GNAT project file, and other source directories with the Emacs project file.

In addition, -P{gpr_file} is added to the project variable gnatmake_opt whenever it is referenced. With the default project variables, this passes the project file to all gnatmake commands.

Lisp variable: ada-prj-default-gpr-file.

link_opt [default: ""]
Holds user linker options; used in the default build commands.

Lisp variable: ada-prj-default-link-opt.

main [default: current file]
Specifies the name of the executable file for the project; used in the default build commands.
make_cmd [default: "${cross_prefix}gnatmake -o ${main} ${main} ${gnatmake_opt} -cargs ${comp_opt} -bargs ${bind_opt} -largs ${link_opt}"]
Command used to build the application.

Lisp variable: ada-prj-default-make-cmd.

obj_dir [default: "."]
A list of directories to search for library files. Ada mode searches this list for the `.ali' files generated by GNAT that contain cross-reference information.

The compiler commands must place the `.ali' files in one of these directories; the default commands do that.

remote_machine [default: ""]
Name of the machine to log into before issuing the compile and build commands. If this variable is empty, the command will be run on the local machine.
run_cmd [default: "./${main}"]
Command used to run the application.
src_dir [default: "."]
A list of directories to search for source files, both for compile commands and source navigation.


Next: , Previous: Project files, Up: Top

6 Compiling Examples

We present several small projects, and walk thru the process of compiling, linking, and running them.

The first example illustrates more Ada mode features than the others; you should work thru that example before doing the others.

All of these examples assume you are using GNAT.

The source for these examples is available on the Emacs Ada mode website mentioned in See Installation.


Next: , Previous: Compiling Examples, Up: Compiling Examples

6.1 No project files

This example uses no project files.

First, create a directory Example_1, containing:

hello.adb:

     with Ada.Text_IO;
     procedure Hello
     is begin
        Put_Line("Hello from hello.adb");
     end Hello;

Yes, this is missing “use Ada.Text_IO;” - we want to demonstrate compiler error handling.

hello_2.adb:

     with Hello_Pkg;
     procedure Hello_2
     is begin
        Hello_Pkg.Say_Hello;
     end Hello_2;

This file has no errors.

hello_pkg.ads:

     package Hello_Pkg is
        procedure Say_Hello;
     end Hello_Pkg;

This file has no errors.

hello_pkg.adb:

     with Ada.Text_IO;
     package Hello_Pkg is
        procedure Say_Hello
        is begin
           Ada.Text_IO.Put_Line ("Hello from hello_pkg.adb");
        end Say_Hello;
     end Hello_Pkg;

Yes, this is missing the keyword body; another compiler error example.

In buffer hello.adb, invoke <Ada | Check file>. You should get a *compilation* buffer containing something like (the directory paths will be different):

     cd c:/Examples/Example_1/
     gnatmake -u -c -gnatc -g c:/Examples/Example_1/hello.adb -cargs -gnatq -gnatQ
     gcc -c -Ic:/Examples/Example_1/ -gnatc -g -gnatq -gnatQ -I- c:/Examples/Example_1/hello.adb
     hello.adb:4:04: "Put_Line" is not visible
     hello.adb:4:04: non-visible declaration at a-textio.ads:264
     hello.adb:4:04: non-visible declaration at a-textio.ads:260
     gnatmake: "c:/Examples/Example_1/hello.adb" compilation error

If you have enabled font-lock, the lines with actual errors (starting with hello.adb) are highlighted, with the file name in red.

Now type <C-x `> (on a PC keyboard, <`> is next to <1>). Or you can click the middle mouse button on the first error line. The compilation buffer scrolls to put the first error on the top line, and point is put at the place of the error in the hello.adb buffer.

To fix the error, change the line to be

         Ada.Text_IO.Put_Line ("hello from hello.adb"):

Now invoke <Ada | Show main>; this displays Ada mode main: hello.

Now (in buffer hello.adb), invoke <Ada | Build>. You are prompted to save the file (if you haven't already). Then the compilation buffer is displayed again, containing:

     cd c:/Examples/Example_1/
     gnatmake -o hello hello -g -cargs -gnatq -gnatQ -bargs  -largs
     gcc -c -g -gnatq -gnatQ hello.adb
     gnatbind -x hello.ali
     gnatlink hello.ali -o hello.exe -g

The compilation has succeeded without errors; hello.exe now exists in the same directory as hello.adb.

Now invoke <Ada | Run>. A *run* buffer is displayed, containing

     Hello from hello.adb
     
     Process run finished

That completes the first part of this example.

Now we will compile a multi-file project. Open the file hello_2.adb, and invoke <Ada | Set main and Build>. This finds an error in hello_pkg.adb:

     cd c:/Examples/Example_1/
     gnatmake -o hello_2 hello_2 -g -cargs -gnatq -gnatQ -bargs  -largs
     gcc -c -g -gnatq -gnatQ hello_pkg.adb
     hello_pkg.adb:2:08: keyword "body" expected here [see file name]
     gnatmake: "hello_pkg.adb" compilation error

This demonstrates that gnatmake finds the files needed by the main program. However, it cannot find files in a different directory, unless you use an Emacs Ada mode project file to specify the other directories; See Set source search path, or a GNAT project file; Use GNAT project file.

Invoke <Ada | Show main>; this displays Ada mode main: hello_2.

Move to the error with <C-x `>, and fix the error by adding body:

     package body Hello_Pkg is

Now, while still in hello_pkg.adb, invoke <Ada | Build>. gnatmake successfully builds hello_2. This demonstrates that Emacs has remembered the main file, in the project variable main, and used it for the Build command.

Finally, again while in hello_pkg.adb, invoke <Ada | Run>. The *run* buffer displays Hello from hello_pkg.adb.

One final point. If you switch back to buffer hello.adb, and invoke <Ada | Run>, hello_2.exe will be run. That is because main is still set to hello_2, as you can see when you invoke <Ada | Project | Edit>.

There are three ways to change main:

  1. Invoke <Ada | Set main and Build>, which sets main to the current file.
  2. Invoke <Ada | Project | Edit>, edit main and main, and click <[save]>
  3. Invoke <Ada | Project | Load>, and load a project file that specifies main


Next: , Previous: No project files, Up: Compiling Examples

6.2 Set compiler options

This example illustrates using an Emacs Ada mode project file to set a compiler option.

If you have files from Example_1 open in Emacs, you should close them so you don't get confused. Use menu <File | Close (current buffer)>.

In directory Example_2, create these files:

hello.adb:

     with Ada.Text_IO;
     procedure Hello
     is begin
        Put_Line("Hello from hello.adb");
     end Hello;

This is the same as hello.adb from Example_1. It has two errors; missing “use Ada.Text_IO;”, and no space between Put_Line and its argument list.

hello.adp:

     comp_opt=-gnatyt

This tells the GNAT compiler to check for token spacing; in particular, there must be a space preceding a parenthesis.

In buffer hello.adb, invoke <Ada | Project | Load...>, and select Example_2/hello.adp.

Then, again in buffer hello.adb, invoke <Ada | Set main and Build>. You should get a *compilation* buffer containing something like (the directory paths will be different):

     cd c:/Examples/Example_2/
     gnatmake -o hello hello -g -cargs -gnatyt  -bargs  -largs
     gcc -c -g -gnatyt hello.adb
     hello.adb:4:04: "Put_Line" is not visible
     hello.adb:4:04: non-visible declaration at a-textio.ads:264
     hello.adb:4:04: non-visible declaration at a-textio.ads:260
     hello.adb:4:12: (style) space required
     gnatmake: "hello.adb" compilation error

Compare this to the compiler output in No project files; the gnatmake option -cargs -gnatq -gnatQ has been replaced by -cargs -gnaty, and an additional error is reported in hello.adb on line 4. This shows that hello.adp is being used to set the compiler options.

Fixing the error, linking and running the code proceed as in No project files.


Next: , Previous: Set compiler options, Up: Compiling Examples

6.3 Set source search path

In this example, we show how to deal with files in more than one directory. We start with the same code as in No project files; create those files (with the errors present)

Create the directory Example_3, containing:

hello_pkg.ads:

     package Hello_Pkg is
        procedure Say_Hello;
     end Hello_Pkg;

hello_pkg.adb:

     with Ada.Text_IO;
     package Hello_Pkg is
        procedure Say_Hello
        is begin
           Ada.Text_IO.Put_Line ("Hello from hello_pkg.adb");
        end Say_Hello;
     end Hello_Pkg;

These are the same files from example 1; hello_pkg.adb has an error on line 2.

In addition, create a directory Example_3/Other, containing these files:

Other/hello_3.adb:

     with Hello_Pkg;
     with Ada.Text_IO; use Ada.Text_IO;
     procedure Hello_3
     is begin
        Hello_Pkg.Say_Hello;
        Put_Line ("From hello_3");
     end Hello_3;

There are no errors in this file.

Other/other.adp:

     src_dir=..
     comp_opt=-I..

Note that there must be no trailing spaces.

In buffer hello_3.adb, invoke <Ada | Project | Load...>, and select Example_3/Other/other.adp.

Then, again in hello_3.adb, invoke <Ada | Set main and Build>. You should get a *compilation* buffer containing something like (the directory paths will be different):

     cd c:/Examples/Example_3/Other/
     gnatmake -o hello_3 hello_3 -g -cargs -I.. -bargs  -largs
     gcc -c -g -I.. hello_3.adb
     gcc -c -I./ -g -I.. -I- C:\Examples\Example_3\hello_pkg.adb
     hello_pkg.adb:2:08: keyword "body" expected here [see file name]
     gnatmake: "C:\Examples\Example_3\hello_pkg.adb" compilation error

Compare the -cargs option to the compiler output in Set compiler options; this shows that other.adp is being used to set the compiler options.

Move to the error with <C-x `>. Ada mode searches the list of directories given by src_dir for the file mentioned in the compiler error message.

Fixing the error, linking and running the code proceed as in No project files.


Next: , Previous: Set source search path, Up: Compiling Examples

6.4 Use GNAT project file

In this example, we show how to use a GNAT project file, with no Ada mode project file.

Create the directory Example_4, containing:

hello_pkg.ads:

     package Hello_Pkg is
        procedure Say_Hello;
     end Hello_Pkg;

hello_pkg.adb:

     with Ada.Text_IO;
     package Hello_Pkg is
        procedure Say_Hello
        is begin
           Ada.Text_IO.Put_Line ("Hello from hello_pkg.adb");
        end Say_Hello;
     end Hello_Pkg;

These are the same files from example 1; hello_pkg.adb has an error on line 2.

In addition, create a directory Example_4/Gnat_Project, containing these files:

Gnat_Project/hello_4.adb:

     with Hello_Pkg;
     with Ada.Text_IO; use Ada.Text_IO;
     procedure Hello_4
     is begin
        Hello_Pkg.Say_Hello;
        Put_Line ("From hello_4");
     end Hello_4;

There are no errors in this file.

Gnat_Project/hello_4.gpr:

     Project Hello_4 is
        for Source_Dirs use (".", "..");
     end Hello_4;

In buffer hello_4.adb, invoke <Ada | Project | Load...>, and select Example_4/Gnat_Project/hello_4.gpr.

Then, again in hello_4.adb, invoke <Ada | Set main and Build>. You should get a *compilation* buffer containing something like (the directory paths will be different):

     cd c:/Examples/Example_4/Gnat_Project/
     gnatmake -o hello_4 hello_4 -Phello_4.gpr -cargs -gnatq -gnatQ -bargs  -largs
     gcc -c -g -gnatyt -gnatq -gnatQ -I- -gnatA c:\Examples\Example_4\Gnat_Project\hello_4.adb
     gcc -c -g -gnatyt -gnatq -gnatQ -I- -gnatA c:\Examples\Example_4\hello_pkg.adb
     hello_pkg.adb:2:08: keyword "body" expected here [see file name]
     gnatmake: "c:\examples\example_4\hello_pkg.adb" compilation error

Compare the gcc options to the compiler output in Set compiler options; this shows that hello_4.gpr is being used to set the compiler options.

Fixing the error, linking and running the code proceed as in No project files.


Previous: Use GNAT project file, Up: Compiling Examples

6.5 Use multiple GNAT project files

In this example, we show how to use multiple GNAT project files, specifying the GNAT project search path in an Ada mode project file.

Create the directory Example_4 as specified in Use GNAT project file.

Create the directory Example_5, containing:

hello_5.adb:

     with Hello_Pkg;
     with Ada.Text_IO; use Ada.Text_IO;
     procedure Hello_5
     is begin
        Hello_Pkg.Say_Hello;
        Put_Line ("From hello_5");
     end Hello_5;

There are no errors in this file.

hello_5.adp:

     ada_project_path=../Example_4/Gnat_Project
     gpr_file=hello_5.gpr

hello_5.gpr:

     with "hello_4";
     Project Hello_5 is
        for Source_Dirs use (".");
        package Compiler is
           for Default_Switches ("Ada") use ("-g", "-gnatyt");
        end Compiler;
     end Hello_5;

In buffer hello_5.adb, invoke <Ada | Project | Load...>, and select Example_5/hello_5.adp.

Then, again in hello_5.adb, invoke <Ada | Set main and Build>. You should get a *compilation* buffer containing something like (the directory paths will be different):

     cd c:/Examples/Example_5/
     gnatmake -o hello_5 hello_5 -Phello_5.gpr -g -cargs -gnatq -gnatQ -bargs  -largs
     gcc -c -g -gnatyt -g -gnatq -gnatQ -I- -gnatA c:\Examples\Example_5\hello_5.adb
     gcc -c -g -gnatyt -g -gnatq -gnatQ -I- -gnatA c:\Examples\Example_4\hello_pkg.adb
     hello_pkg.adb:2:08: keyword "body" expected here [see file name]
     gnatmake: "c:\examples\example_4\hello_pkg.adb" compilation error

Now type <C-x `>. Example_4/hello_pkg.adb is shown, demonstrating that hello_5.gpr and hello_4.gpr are being used to set the compilation search path.


Next: , Previous: Compiling Examples, Up: Top

7 Moving Through Ada Code

There are several easy to use commands to navigate through Ada code. All these functions are available through the Ada menu, and you can also use the following key bindings or the command names. Some of these menu entries are available only if the GNAT compiler is used, since the implementation relies on the GNAT cross-referencing information.

M-C-e
Move to the next function/procedure/task, which ever comes next (ada-next-procedure).
M-C-a
Move to previous function/procedure/task (ada-previous-procedure).
M-x ada-next-package
Move to next package.
M-x ada-previous-package
Move to previous package.
C-c C-a
Move to matching start of end (ada-move-to-start). If point is at the end of a subprogram, this command jumps to the corresponding begin if the user option ada-move-to-declaration is nil (default), otherwise it jumps to the subprogram declaration.
C-c C-e
Move point to end of current block (ada-move-to-end).
C-c o
Switch between corresponding spec and body file (ff-find-other-file). If point is in a subprogram, position point on the corresponding declaration or body in the other file.
C-c c-d
Move from any reference to its declaration, for from a declaration to its body (for procedures, tasks, private and incomplete types).
C-c C-r
Runs the gnatfind command to search for all references to the identifier surrounding point (ada-find-references). Use C-x ` (next-error) to visit each reference (as for compilation errors).

If the ada-xref-create-ali variable is non-nil, Emacs will try to run GNAT for you whenever cross-reference information is needed, and is older than the current source file.


Next: , Previous: Moving Through Ada Code, Up: Top

8 Identifier completion

Emacs and Ada mode provide two general ways for the completion of identifiers. This is an easy way to type faster: you just have to type the first few letters of an identifiers, and then loop through all the possible completions.

The first method is general for Emacs. It works by parsing all open files for possible completions.

For instance, if the words `my_identifier', `my_subprogram' are the only words starting with `my' in any of the opened files, then you will have this scenario:

     You type:  my<M-/>
     Emacs inserts:  `my_identifier'
     If you press <M-/> once again, Emacs replaces `my_identifier' with
     `my_subprogram'.
     Pressing <M-/> once more will bring you back to `my_identifier'.

This is a very fast way to do completion, and the casing of words will also be respected.

The second method (<C-TAB>) is specific to Ada mode and the GNAT compiler. Emacs will search the cross-information for possible completions.

The main advantage is that this completion is more accurate: only existing identifier will be suggested.

On the other hand, this completion is a little bit slower and requires that you have compiled your file at least once since you created that identifier.

C-<TAB>
Complete current identifier using cross-reference information.
M-/
Complete identifier using buffer information (not Ada-specific).


Next: , Previous: Identifier completion, Up: Top

9 Automatic Smart Indentation

Ada mode comes with a full set of rules for automatic indentation. You can also configure the indentation, via the following variables:

ada-broken-indent (default value: 2)
Number of columns to indent the continuation of a broken line.
ada-indent (default value: 3)
Number of columns for default indentation.
ada-indent-record-rel-type (default value: 3)
Indentation for record relative to type or use.
ada-indent-return (default value: 0)
Indentation for return relative to function (if ada-indent-return is greater than 0), or the open parenthesis (if ada-indent-return is negative or 0). Note that in the second case, when there is no open parenthesis, the indentation is done relative to function with the value of ada-broken-indent.
ada-label-indent (default value: -4)
Number of columns to indent a label.
ada-stmt-end-indent (default value: 0)
Number of columns to indent a statement end keyword on a separate line.
ada-when-indent (default value: 3)
Indentation for when relative to exception or case.
ada-indent-is-separate (default value: t)
Non-nil means indent is separate or is abstract if on a single line.
ada-indent-to-open-paren (default value: t)
Non-nil means indent according to the innermost open parenthesis.
ada-indent-after-return (default value: t)
Non-nil means that the current line will also be re-indented before inserting a newline, when you press <RET>.

Most of the time, the indentation will be automatic, i.e when you press <RET>, the cursor will move to the correct column on the next line.

You can also indent single lines, or the current region, with <TAB>.

Another mode of indentation exists that helps you to set up your indentation scheme. If you press C-c <TAB>, Ada mode will do the following:

The exact indentation of the current line is the same as the one for the reference line, plus an offset given by the variable.

<TAB>
Indent the current line or the current region.
C-M-\
Indent lines in the current region.
C-c <TAB>
Indent the current line and display the name of the variable used for indentation.


Next: , Previous: Automatic Smart Indentation, Up: Top

10 Formatting Parameter Lists

C-c C-f
Format the parameter list (ada-format-paramlist).

This aligns the declarations on the colon (`:') separating argument names and argument types, and aligns the in, out and in out keywords.


Next: , Previous: Formatting Parameter Lists, Up: Top

11 Automatic Casing

Casing of identifiers, attributes and keywords is automatically performed while typing when the variable ada-auto-case is set. Every time you press a word separator, the previous word is automatically cased.

You can customize the automatic casing differently for keywords, attributes and identifiers. The relevant variables are the following: ada-case-keyword, ada-case-attribute and ada-case-identifier.

All these variables can have one of the following values:

downcase-word
The word will be lowercase. For instance My_vARIable is converted to my_variable.
upcase-word
The word will be uppercase. For instance My_vARIable is converted to MY_VARIABLE.
ada-capitalize-word
The first letter and each letter following an underscore (`_') are uppercase, others are lowercase. For instance My_vARIable is converted to My_Variable.
ada-loose-case-word
Characters after an underscore `_' character are uppercase, others are not modified. For instance My_vARIable is converted to My_VARIable.

Ada mode allows you to define exceptions to these rules, in a file specified by the variable variable ada-case-exception-file (default ~/.emacs_case_exceptions). Each line in this file specifies the casing of one word or word fragment. Comments may be included, separated from the word by a space.

If the word starts with an asterisk (<*>), it defines the casing af a word fragemnt (or “substring”); part of a word between two underscores or word boundary.

For example:

     DOD        Department of Defense
     *IO
     GNAT       The GNAT compiler from Ada Core Technologies

The word fragment *IO applies to any word containing “_io”; Text_IO, Hardware_IO, etc.

There are two ways to add new items to this file: you can simply edit it as you would edit any text file. Or you can position point on the word you want to add, and select menu `Ada | Edit | Create Case Exception', or press C-c C-y (ada-create-case-exception). The word will automatically be added to the current list of exceptions and to the file.

To define a word fragment case exception, select the word fragment, then select menu `Ada | Edit | Create Case Exception Substring'.

It is sometimes useful to have multiple exception files around (for instance, one could be the standard Ada acronyms, the second some company specific exceptions, and the last one some project specific exceptions). If you set up the variable ada-case-exception-file as a list of files, each of them will be parsed and used in your emacs session. However, when you save a new exception through the menu, as described above, the new exception will be added to the first file in the list.

C-c C-b
Adjust case in the whole buffer (ada-adjust-case-buffer).
C-c C-y
Create a new entry in the exception dictionary, with the word under the cursor (ada-create-case-exception)
C-c C-t
Rereads the exception dictionary from the file ada-case-exception-file (ada-case-read-exceptions).


Next: , Previous: Automatic Casing, Up: Top

12 Statement Templates

Templates are defined for most Ada statements, using the Emacs “skeleton” package. They can be inserted in the buffer using the following commands:

C-c t b
exception Block (ada-exception-block).
C-c t c
case (ada-case).
C-c t d
declare Block (ada-declare-block).
C-c t e
else (ada-else).
C-c t f
for Loop (ada-for-loop).
C-c t h
Header (ada-header).
C-c t i
if (ada-if).
C-c t k
package Body (ada-package-body).
C-c t l
loop (ada-loop).
C-c p
subprogram body (ada-subprogram-body).
C-c t t
task Body (ada-task-body).
C-c t w
while Loop (ada-while).
C-c t u
use (ada-use).
C-c t x
exit (ada-exit).
C-c t C-a
array (ada-array).
C-c t C-e
elsif (ada-elsif).
C-c t C-f
function Spec (ada-function-spec).
C-c t C-k
package Spec (ada-package-spec).
C-c t C-p
procedure Spec (ada-package-spec.
C-c t C-r
record (ada-record).
C-c t C-s
subtype (ada-subtype).
C-c t C-t
task Spec (ada-task-spec).
C-c t C-u
with (ada-with).
C-c t C-v
private (ada-private).
C-c t C-w
when (ada-when).
C-c t C-x
exception (ada-exception).
C-c t C-y
type (ada-type).


Next: , Previous: Statement Templates, Up: Top

13 Comment Handling

By default, comment lines get indented like Ada code. There are a few additional functions to handle comments:

M-;
Start a comment in default column.
M-j
Continue comment on next line.
C-c ;
Comment the selected region (add – at the beginning of lines).
C-c :
Uncomment the selected region
M-q
autofill the current comment.


Previous: Comment Handling, Up: Top

Index

Table of Contents