summaryrefslogtreecommitdiff
path: root/libs/cglm/docs/source/opengl.rst
diff options
context:
space:
mode:
authorsanine <sanine.not@pm.me>2022-06-14 00:06:42 -0500
committersanine <sanine.not@pm.me>2022-06-14 00:06:42 -0500
commit2f518e5e28d35ae24a5ac0e31000835e43b01972 (patch)
tree47fdeb9fa5b04e267702acb06424d3f87b37dd84 /libs/cglm/docs/source/opengl.rst
parent034d5c965ff34cfdf4b153af9f32360a02e35684 (diff)
add cglm as 3rd-party library
Diffstat (limited to 'libs/cglm/docs/source/opengl.rst')
-rw-r--r--libs/cglm/docs/source/opengl.rst61
1 files changed, 61 insertions, 0 deletions
diff --git a/libs/cglm/docs/source/opengl.rst b/libs/cglm/docs/source/opengl.rst
new file mode 100644
index 0000000..2cebae0
--- /dev/null
+++ b/libs/cglm/docs/source/opengl.rst
@@ -0,0 +1,61 @@
+How to send vector or matrix to OpenGL like API
+==================================================
+
+*cglm*'s vector and matrix types are arrays. So you can send them directly to a
+function which accepts pointer. But you may got warnings for matrix because it is
+two dimensional array.
+
+Passing / Uniforming Matrix to OpenGL:
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+**glUniformMatrix4fv** accepts float pointer, you can pass matrix to that parameter
+and it should work but with warnings. "You can pass" doesn't mean that you must pass like that.
+
+**Correct options:**
+
+Correct doesn't mean correct way to use OpenGL it is just shows correct way to pass cglm type to it.
+
+1. Pass first column
+---------------------
+
+The goal is that pass address of matrix, first element of matrix is also address of matrix,
+because it is array of vectors and vector is array of floats.
+
+.. code-block:: c
+
+ mat4 matrix;
+ /* ... */
+ glUniformMatrix4fv(location, 1, GL_FALSE, matrix[0]);
+
+array of matrices:
+
+.. code-block:: c
+
+ mat4 matrix;
+ /* ... */
+ glUniformMatrix4fv(location, count, GL_FALSE, matrix[0][0]);
+
+1. Cast matrix to pointer
+--------------------------
+
+.. code-block:: c
+
+ mat4 matrix;
+ /* ... */
+ glUniformMatrix4fv(location, count, GL_FALSE, (float *)matrix);
+
+in this way, passing aray of matrices is same
+
+Passing / Uniforming Vectors to OpenGL:
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+You don't need to do extra thing when passing cglm vectors to OpengL or other APIs.
+Because a function like **glUniform4fv** accepts vector as pointer. cglm's vectors
+are array of floats. So you can pass it directly ot those functions:
+
+.. code-block:: c
+
+ vec4 vec;
+ /* ... */
+ glUniform4fv(location, 1, vec);
+
+this show how to pass **vec4** others are same.