{ "version": "https:\/\/jsonfeed.org\/version\/1.1", "title": "Guides", "_rss_description": "A blog by George Mishurovsky — a senior software engineer with a medical degree. Drawing from both engineering and scientific thinking, he explores software, architecture, design, psychology, and product thinking.", "_rss_language": "en", "_itunes_email": "george@mishurovsky.com", "_itunes_categories_xml": "", "_itunes_image": "https:\/\/mishurovsky.com\/blog\/pictures\/userpic\/userpic-square@2x.jpg?1753619610", "_itunes_explicit": "no", "home_page_url": "https:\/\/mishurovsky.com\/blog\/?go=tags\/guides\/", "feed_url": "https:\/\/mishurovsky.com\/blog\/?go=tags%2Fguides%2Fjson%2F", "icon": "https:\/\/mishurovsky.com\/blog\/pictures\/userpic\/userpic@2x.jpg?1753619610", "authors": [ { "name": "George Mishurovsky", "url": "https:\/\/mishurovsky.com\/blog\/", "avatar": "https:\/\/mishurovsky.com\/blog\/pictures\/userpic\/userpic@2x.jpg?1753619610" } ], "items": [ { "id": "17", "url": "https:\/\/mishurovsky.com\/blog\/?go=all\/how-to-bulk-rename-some-entity-and-files-in-a-project\/", "title": "Renaming Entities Project-Wide with find, grep, sed, and rename", "content_html": "

There are times in software projects when a big shift happens in domain representation. This results in changes of project structure, class responsibilities, and occasionally, requires bulk renames of entities across the whole codebase.<\/p>\n

\n\"\"\n<\/div>\n

Imagine we need to rename every Employee<\/span> to Worker<\/span>. This change should affect both file paths and textual occurrences throughout the project.<\/p>\n

Renaming may sound like a simple problem: assuming there are no external dependencies using the target name, we just need to rename all occurrences of it inside a repository. But there multiple caveats:<\/p>\n

    \n
  1. We must rename both file names and folder names.<\/li>\n
  2. We must rename all code and text occurrences.<\/li>\n
  3. Casing must be preserved: Employee → Worker<\/span>, and employee → worker<\/span>.<\/li>\n
  4. Both plain and compound usages must be properly renamed: createEmployee → createWorker<\/span>.<\/li>\n
  5. Non-code, non-document files must not be affected (consider binary files which by coincidence might have an ...employee...<\/span> fragment inside).<\/li>\n
  6. There are folders or files which we would want to omit from renaming (e. g., .git<\/span>).<\/li>\n
  7. No IDE provides such functionality, so we cannot rely on existing solutions.<\/li>\n<\/ol>\n

    I will address all these challenges in a solution below, but there is an important complexity that cannot be tackled with automation. If by any chance your code depends on a library with the target name (e. g., employee.js<\/span>) or uses exports from a library containing the target name (import type { valuableEmployeee } from ‘employee-js’<\/span>), you’ll have to resolve issues manually after renaming.<\/p>\n

    Reviewing Expected Changes<\/h2>\n

    First, remove any folders and files that are recreated during project builds or setup: build\/<\/span>, dist\/<\/span>, node_modules\/<\/span>, .storybook-static\/<\/span>, etc. This step isn’t strictly necessary, but it can help iterate faster if commands encounter errors.<\/p>\n

    Now, let’s start with renaming text occurrences by listing all files that might get affected with find<\/span> command. Here I am using -iname<\/span> for case-insensitive search and `-not -path ‘’` syntax to exclude folders we need to protect from changes.<\/p>\n

    Shell<\/span>