improved chunking pipeline and handling of expert settings

This commit is contained in:
PaulKoudelka committed 2026-07-29 18:47:59 +02:00
1 parent 2829051e4c
commit 89f7156fdf
16 files changed
+764 -138

No files matched your search

@@ -3817,6 +3817,12 @@ UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCEERIV1USERNAMEPASSWORDEXPORTDIALOG:
-- Cancel
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCEERIV1USERNAMEPASSWORDEXPORTDIALOG::T900713019"] = "Cancel"
-- Hide Expert Settings
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T1108876344"] = "Hide Expert Settings"
-- Optional expert settings for how this data source is split before embedding.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T1133561850"] = "Optional expert settings for how this data source is split before embedding."
-- Describe what data this directory contains to help the AI select it.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T1136409150"] = "Describe what data this directory contains to help the AI select it."
@@ -3826,6 +3832,9 @@ UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T1265737624"
-- Selected base directory for this data source
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T1312296210"] = "Selected base directory for this data source"
-- Maximum number of tokens per chunk for this data source. Use 0 to use the embedding provider setting.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T1529596777"] = "Maximum number of tokens per chunk for this data source. Use 0 to use the embedding provider setting."
-- Description
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T1725856265"] = "Description"
@@ -3838,24 +3847,48 @@ UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T1847791252"
-- In order for the AI to be able to determine the appropriate data at any time, you must choose an embedding method.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T1948697886"] = "In order for the AI to be able to determine the appropriate data at any time, you must choose an embedding method."
-- The data source max chunk length must be less than or equal to the embedding provider limit ({0}).
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T2166221195"] = "The data source max chunk length must be less than or equal to the embedding provider limit ({0})."
-- Number of tokens repeated at the start of the next chunk. Use 0 to disable overlap.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T2263831524"] = "Number of tokens repeated at the start of the next chunk. Use 0 to disable overlap."
-- Add
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T2646845972"] = "Add"
-- Chunk overlap length
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T2801813827"] = "Chunk overlap length"
-- The embedding you selected runs locally or in your organization. Your data is not sent to the cloud.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T2814869210"] = "The embedding you selected runs locally or in your organization. Your data is not sent to the cloud."
-- Embedding
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T2838542994"] = "Embedding"
-- Please enter 0 or a positive max chunk length.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T3130373769"] = "Please enter 0 or a positive max chunk length."
-- Please enter 0 or a positive overlap length.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T3242265813"] = "Please enter 0 or a positive overlap length."
-- For some data types, such as Office files, MindWork AI Studio requires the open-source application Pandoc.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T3359366900"] = "For some data types, such as Office files, MindWork AI Studio requires the open-source application Pandoc."
-- Show Expert Settings
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T3361153305"] = "Show Expert Settings"
-- Max chunk length
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T3621670500"] = "Max chunk length"
-- Your security policy
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T4081226330"] = "Your security policy"
-- Select the base directory
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T562479068"] = "Select the base directory"
-- The overlap must be smaller than the effective max chunk length.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T57492807"] = "The overlap must be smaller than the effective max chunk length."
-- Data Source Name
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYDIALOG::T813773421"] = "Data Source Name"
@@ -3934,9 +3967,18 @@ UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYINFODIALOG::T4438734
-- The directory chosen for the data source exists.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALDIRECTORYINFODIALOG::T445858624"] = "The directory chosen for the data source exists."
-- Hide Expert Settings
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T1108876344"] = "Hide Expert Settings"
-- Optional expert settings for how this data source is split before embedding.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T1133561850"] = "Optional expert settings for how this data source is split before embedding."
-- Select a file for this data source. The content of this file will be processed for the data source.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T1190880267"] = "Select a file for this data source. The content of this file will be processed for the data source."
-- Maximum number of tokens per chunk for this data source. Use 0 to use the embedding provider setting.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T1529596777"] = "Maximum number of tokens per chunk for this data source. Use 0 to use the embedding provider setting."
-- Description
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T1725856265"] = "Description"
@@ -3949,9 +3991,18 @@ UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T1847791252"] = "
-- In order for the AI to be able to determine the appropriate data at any time, you must choose an embedding method.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T1948697886"] = "In order for the AI to be able to determine the appropriate data at any time, you must choose an embedding method."
-- The data source max chunk length must be less than or equal to the embedding provider limit ({0}).
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T2166221195"] = "The data source max chunk length must be less than or equal to the embedding provider limit ({0})."
-- Number of tokens repeated at the start of the next chunk. Use 0 to disable overlap.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T2263831524"] = "Number of tokens repeated at the start of the next chunk. Use 0 to disable overlap."
-- Add
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T2646845972"] = "Add"
-- Chunk overlap length
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T2801813827"] = "Chunk overlap length"
-- The embedding you selected runs locally or in your organization. Your data is not sent to the cloud.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T2814869210"] = "The embedding you selected runs locally or in your organization. Your data is not sent to the cloud."
@@ -3961,15 +4012,30 @@ UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T2838542994"] = "
-- Describe what data this file contains to help the AI select it.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T2859265837"] = "Describe what data this file contains to help the AI select it."
-- Please enter 0 or a positive max chunk length.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T3130373769"] = "Please enter 0 or a positive max chunk length."
-- Please enter 0 or a positive overlap length.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T3242265813"] = "Please enter 0 or a positive overlap length."
-- For some data types, such as Office files, MindWork AI Studio requires the open-source application Pandoc.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T3359366900"] = "For some data types, such as Office files, MindWork AI Studio requires the open-source application Pandoc."
-- Show Expert Settings
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T3361153305"] = "Show Expert Settings"
-- Max chunk length
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T3621670500"] = "Max chunk length"
-- Select the file
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T3740148848"] = "Select the file"
-- Your security policy
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T4081226330"] = "Your security policy"
-- The overlap must be smaller than the effective max chunk length.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T57492807"] = "The overlap must be smaller than the effective max chunk length."
-- Data Source Name
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::DATASOURCELOCALFILEDIALOG::T813773421"] = "Data Source Name"
@@ -4132,12 +4198,18 @@ UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGMETHODDIALOG::T662524223"] = "A lin
-- Cancel
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGMETHODDIALOG::T900713019"] = "Cancel"
-- Hide Expert Settings
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T1108876344"] = "Hide Expert Settings"
-- Failed to store the API key in the operating system. The message was: {0}. Please try again.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T1122745046"] = "Failed to store the API key in the operating system. The message was: {0}. Please try again."
-- API Key
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T1324664716"] = "API Key"
-- Please be aware: This section is for experts only. For cloud providers, the selected tokenizer and chunk settings may not match the real embedding model limits exactly.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T1345053261"] = "Please be aware: This section is for experts only. For cloud providers, the selected tokenizer and chunk settings may not match the real embedding model limits exactly."
-- Create account
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T1356621346"] = "Create account"
@@ -4159,18 +4231,21 @@ UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T1847791252"] = "Up
-- Failed to load the API key from the operating system. The message was: {0}. You might ignore this message and provide the API key again.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T1870831108"] = "Failed to load the API key from the operating system. The message was: {0}. You might ignore this message and provide the API key again."
-- Maximum number of tokens sent to the embedding model per chunk. The default is 16,384.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T2064265557"] = "Maximum number of tokens sent to the embedding model per chunk. The default is 16,384."
-- Model
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T2189814010"] = "Model"
-- Embedding batch size
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T2209963239"] = "Embedding batch size"
-- (Optional) API Key
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T2331453405"] = "(Optional) API Key"
-- Invalid tokenizer:
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T2448302543"] = "Invalid tokenizer:"
-- Maximum number of tokens sent to the embedding model per chunk.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T2544519277"] = "Maximum number of tokens sent to the embedding model per chunk."
-- Add
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T2646845972"] = "Add"
@@ -4192,12 +4267,15 @@ UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T2961294165"] = "To
-- Please enter a token limit greater than 0.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T3316544737"] = "Please enter a token limit greater than 0."
-- Show Expert Settings
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T3361153305"] = "Show Expert Settings"
-- How many chunks are sent to the embedding provider at once. The default is 1.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T3780233303"] = "How many chunks are sent to the embedding provider at once. The default is 1."
-- Choose a custom tokenizer here
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T3787466119"] = "Choose a custom tokenizer here"
-- For better embeddings and less storage usage, it's recommended to use a custom tokenizer to enable a more accurate token count.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T4126312157"] = "For better embeddings and less storage usage, it's recommended to use a custom tokenizer to enable a more accurate token count."
-- Model selection
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T416738168"] = "Model selection"
@@ -4207,6 +4285,9 @@ UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T504465522"] = "We
-- Host
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T808120719"] = "Host"
-- Please enter an embedding batch size greater than 0.
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T840259907"] = "Please enter an embedding batch size greater than 0."
-- Provider
UI_TEXT_CONTENT["AISTUDIO::DIALOGS::EMBEDDINGPROVIDERDIALOG::T900237532"] = "Provider"