fix: Ensure parser directory is created correctly, not as a file

[tomoya0318]

Overview

This PR fixes a critical build failure where the build script incorrectly created the parser build directory as a file instead of a directory.

This change ensures the directory is proactively created correctly, making the build process robust. It also includes minor code style improvements by removing unnecessary whitespace.

Background/Problem

The core issue was not simply that the parser directory was missing, but that the build script incorrectly created it as a file.

Here's the sequence of events that caused the failure:

1. The cp command within the cpResources.sh script was executed with a non-existent destination path (MSCCD/modules/msccd_tokenizer/src/main/java/org/nagoya_u/ertl/sa/parser).
2. Instead of creating a directory, the cp command's default behavior created a file named parser containing the contents of the first source file.
3. All subsequent attempts to copy other files into what should have been a directory immediately failed with a target is not a directory error.

Changes

To resolve this, the following changes were made:

• Proactive Directory Creation: In tokenizerGeneration.py, added a call to os.makedirs() right before the resource copy script is executed. This proactively creates parser as a directory, ensuring that the cp command correctly copies files into it, rather than creating an incorrect file.
• Code Style: Removed unnecessary trailing whitespace from the script for better readability and consistency.

How to Test

I have confirmed that this change fixes the issue by following these steps:

1. Checked out this branch.
2. Manually removed the parser directory to simulate the failure condition.
3. Executed the generation script

python3 tokenizerGeneration.py

4. Result: The script now completes successfully without any errors, and the tokenizer .jar file is generated as expected.

Related Issues

• N/A