Build a text formula from real delimiters without assuming every row has the same shape. Fournir: Text to extract, Observed input patterns, Excel version and rules. Résultat attendu: A robust extraction formula with malformed-input checks.
Votre texte reste dans ce navigateur. AILesson Prompts ne l’envoie ni à un modèle ni à un serveur.
Les champs non remplis restent visibles sous forme d’espaces réservés, afin que vous puissiez quand même copier et modifier le prompt.
ChatGPTCreate an Excel formula to extract text.
Desired result:
[goal]
Observed input patterns:
[patterns]
Excel version and exception rules:
[version]
Identify the delimiter or positional rule and test whether it is stable across the supplied values. Do not assume all names have two words or all codes have fixed length unless stated. Return a modern formula, a legacy-compatible alternative when relevant, and tests for missing delimiters, repeated delimiters, leading/trailing spaces, blanks, and non-ASCII text. Keep malformed values visibly flagged rather than silently truncating them.De l'entrée au résultat
Voyez comment un contexte concret transforme cette recette en résultat utilisable.
The stable rule is text between one “<” and the following “>”. Because repeated delimiters are considered malformed, count them before extracting.
B2 formula
=IF(TRIM(A2)="","",IF(AND(LEN(A2)-LEN(SUBSTITUTE(A2,"<",""))=1,LEN(A2)-LEN(SUBSTITUTE(A2,">",""))=1),TRIM(TEXTBEFORE(TEXTAFTER(A2,"<"),">")),"Review"))
The two length differences count opening and closing brackets. Only one of each allows extraction; TRIM removes surrounding spaces without changing accented or Chinese characters.
Legacy approach: use MID with FIND after the same bracket-count checks: TRIM(MID(A2,FIND("<",A2)+1,FIND(">",A2)-FIND("<",A2)-1)).
Tests: Ana Pérez and 王小明 return their full email addresses; blank returns blank; missing brackets and two opening brackets return Review. Also verify that “>” occurs after “<”; if reversed brackets exist in the data, add that condition before rollout.
Observed patterns prevent a formula designed for one clean example from corrupting irregular rows.
A visible malformed-input result makes later cleanup auditable.
Does the extraction rule rely on a delimiter that is actually stable?
Are missing or duplicated delimiters reported rather than misparsed?
Does the formula preserve accents and non-Latin characters?
Utilisez-la en toute confiance
Des réponses pratiques sur le bon moment pour utiliser cette recette, ce qu’il faut fournir et les cas où une vérification humaine reste nécessaire.
For “Extract names, emails, or codes in Excel,” prepare Text to extract, Observed input patterns, and Excel version and rules. Replace placeholders only with information you can verify. If a detail is unknown, preserve that uncertainty explicitly instead of asking the model to infer it.
The result is not ready if it does not yet deliver the stated outcome—A robust extraction formula with malformed-input checks—from the supplied evidence, or if it relies on unresolved assumptions, missing approvals, or invented details. Use the checks as release gates: revise the source inputs or assign a named, authorized reviewer instead of polishing an unsupported output.
The published test record for “Extract names, emails, or codes in Excel” lists ChatGPT as of 2026-08-28. This confirms recorded runs, not guaranteed compatibility or identical results in later product versions. For another tool or version, keep every constraint visible and repeat the result checks before use.
Plus de façons d'explorer
Faites avancer votre travail
Analyses de tableurs
Convert verified metrics into decisions, changes, risks, and requests without overstating evidence
Ouvrir la recetteAnalyses de tableurs
Separate names, addresses, or codes using verified delimiters and exception flags
Ouvrir la recetteAnalyses de tableurs
Map known label variants to an approved vocabulary while exposing unknown values
Ouvrir la recetteAILesson · Cours recommandés
Passez de la compréhension de l'IA à l'action. Entraînez-vous à formuler des requêtes, à évaluer et à améliorer les résultats grâce à des leçons interactives pour le travail et la vie quotidienne.
15 leçons interactives
Lancez une conversation utile avec une IA, améliorez sa réponse, vérifiez l’essentiel et sachez quand c’est à vous d’agir.
Découvrir ce cours
10 leçons interactives
Découvrez ChatGPT à travers trois flux de travail pratiques : mener une première conversation, travailler à partir d’un fichier, puis utiliser Projects pour réutiliser un plan vérifié dans une nouvelle conversation.
Découvrir ce cours
7 leçons interactives
Comprenez comment les systèmes courants de chat IA apprennent des structures linguistiques, utilisent les messages actuels, construisent leurs réponses et intègrent des sources et des outils.
Découvrir ce cours