| Invelos Forums->Posts by Clark G. Flipper |
Page:
1  Previous Next
|
| Message |
Details |
For invidual movies I got this copilot thing working now (not yet for serials). Use the following prompt (adaping the movie name):
You are a precise web‑scraping and data‑transformation agent.
**TASK**
Retrieve the IMDb page for the film **“Mein Schulfreund” (1960), directed by Robert Siodmak**.
Navigate to the **Full Cast & Crew** section and extract **all crew credits**, explicitly **excluding all cast/acting information**.
---
## **DATA EXTRACTION RULES**
### **1. Person Name Parsing**
For each crew member, split the IMDb‑listed name into:
- **FirstName**
- **MiddleName**
- **LastName**
Rules:
- If the name has two parts → first = FirstName, second = LastName, MiddleName = `""`.
- If the name has more than two parts →
- First token → FirstName
- Last token → LastName
- All tokens in between → MiddleName (space‑joined)
- Initials remain as written (e.g., “A.” stays “A.”).
---
### **2. Birth Year**
- If IMDb provides a birth year on the person’s profile → use that integer.
- If no birth year is available → use **0**.
---
### **3. CreditType and CreditSubtype Mapping**
Assign each credit to one of the following **CreditType** values:
- Direction
- Writing
- Production
- Cinematography
- Film Editing
- Music
- Sound
- Art
- Other
Then assign the appropriate **CreditSubtype**:
**Direction:** Director | Custom
**Writing:** Original Material by | Screenwriter | Writer | Original Characters | Created by | Story by | Developed by | Custom
**Production:** Producer | Executive Producer | Custom
**Cinematography:** Director of Photography | Cinematographer | Custom
**Film Editing:** Film Editor | Custom
**Music:** Composer | Song Writer | Theme by | Custom
**Sound:** Sound | Sound Designer | Supervising Sound Editor | Sound Editor | Sound Re‑Recording Mixer | Production Sound Mixer | Custom
**Art:** Production Designer | Art Director | Costume Designer | Make‑up Artist | Visual Effects | Make‑up Effects | Creature Design | Custom
**Other:** Custom
---
### **4. Optional Attributes**
- **CreditedAs** → include only if IMDb lists an alternative spelling (e.g., “as J.M. Simmel”).
- **CustomRole** → include only when:
- The role is not covered by the allowed subtypes, or
- The subtype is “Custom”.
Examples:
- “assistant director”
- “script consultant”
- “voice actor – dubbing (Italian)”
---
## **OUTPUT REQUIREMENTS**
Produce **one XML document** that **validates exactly** against the following XSD:
```xml
<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema" elementFormDefault="qualified">
<xs:element name="CrewInformation" type="CrewInformation"/>
<xs:complexType name="CrewInformation">
<xs:sequence>
<xs:element name="Title" type="xs:string" minOccurs="1"/>
<xs:element name="Credits" type="Credits"/>
</xs:sequence>
</xs:complexType>
<xs:complexType name="Credit">
<xs:attribute name="FirstName" type="xs:string" use="required"/>
<xs:attribute name="MiddleName" type="xs:string" use="required"/>
<xs:attribute name="LastName" type="xs:string" use="required"/>
<xs:attribute name="BirthYear" type="xs:int"/>
<xs:attribute name="CreditType" type="xs:string" use="required"/>
<xs:attribute name="CreditSubtype" type="xs:string" use="required"/>
<xs:attribute name="CreditedAs" type="xs:string" use="optional"/>
<xs:attribute name="CustomRole" type="xs:string" use="optional"/>
</xs:complexType>
<xs:complexType name="Credits">
<xs:choice minOccurs="0" maxOccurs="unbounded">
<xs:element name="Credit" type="Credit"/>
</xs:choice>
</xs:complexType>
</xs:schema>
```
### **XML Structure**
```xml
<CrewInformation>
<Title>Mein Schulfreund – Crew (IMDb)</Title>
<Credits>
<!-- multiple <Credit .../> elements -->
</Credits>
</CrewInformation>
```
---
## **GOAL**
Produce a **complete, lossless, XSD‑valid XML representation** of all crew credits for *Mein Schulfreund* (1960), with correct name parsing, birth‑year resolution, and role classification. | Posted: Topic Replies: 776, Topic Views: 71781 |  | I suggest to use AI for this task. With Copilot you can do something like:
Retrieve the first season of the TV series Gunsmoke from IMDb, access all episodes of that season, and extract the available crew information for each episode (especially directing and writing credits). Then generate a structured XML file that includes:
• a series header with the overall series crew (creators, main cast)
• an individual <episode> block for each of the 39 episodes
• the episode‑specific crew data as listed on IMDb
• meaningful attributes such as number, title, and imdb_id
The resulting XML file should be complete, consistent, and machine‑readable. | Posted: Topic Replies: 776, Topic Views: 71781 | |
|
| Invelos Forums->Posts by Clark G. Flipper |
Page:
1 Previous Next
|