Hi Christian,

welcome to the Oracle Analytics community and forum.

What product and version are you working with?

Because you are running Catalog Manager on linux, it looks like you are on-premises (OBIEE or OAS).

All the OBIEE versions have the catalog on disk, allowing you to do a full text search by using linux commands on the files on disk. This would let you find all the places where the replace will happen, and maybe adjust your replace syntax. Sometimes doing 2-3 replaces works better, each time you work on a smaller subset of matches to avoid the wrong matches like you described.

If you are on OAS, the catalog is in the database, you could do the same search via a SQL query (I would still use the Catalog Manager tool to do the replace if you aren't familiar with the details of how the catalog objects are encoded in the database, the various hashes etc.).

Both these options cover your Q2: in Catalog Manager there isn't just a "search" option, therefore you can't easily get the list of objects (maybe you could get close to it by doing a report, to get all objects code and then filter those by searching your text etc.).

For Q1, not really: the Catalog Manager does have that "basic" search & replace. It does work, but it has limited options and control. You can define some extra details on what you want to replace if you use a file containing the various search and replace options, but if you really want to replace a piece of text (and not a column etc.), you don't have many options.

Some info to get you started with the OAS catalog in the database.

Everything is in the BIPLATFORM database schema.

The table css_si_files contains the list of files in the catalog (and other things as well), filter by column namespace = 'content:catalog' to have only catalog objects. You can use full_path to have the absolute path of each object, and content_id is the foreign key to the table with the catalog objects content.

The files content is in the table css_file_content, the key is content_id, that you get from the previous table.
The content of the file is in the column data, but you need to pay attention to the column compressed_type: if it's 'gzip', it means that the data is compressed and you need to uncompress it first.
In an Oracle database you can use utl_compress.lz_uncompress(data) to uncompress the value.
On the data column you can make SQL queries because it either contains XML or JSON, and the database "speaks" both these formats (for some other files it can be binary content, but you aren't meant to play with those files anyway, anything in /system you better stay away ;-) )

Remember: it is not supported to edit the database directly (can be done, but you better be sure about what you are doing). But you can freely use it in a read-only approach to find interesting things etc. For example you can find all the matches that will be replaced, and from there either have a list of things to fix after the replace, or be able to look into optimizing your search & replace strategy.